Search in the document preview
Immigrants in the Educational System in Spain: Who Persists?
Verónica de Miguel-Luken1 • Miguel Solana-Solana2
Accepted: 28 March 2016 / Published online: 16 April 2016 Springer Science+Business Media Dordrecht 2016
Abstract Educational attainment in one’s youth generally leads to a more successful transition to adulthood. However, the number of years that youth spend in the educational
system is affected by personal characteristics and social context. For immigrants who
arrive in a country as children, one of these variables is national or geographical origin,
which influences the type of adjustment challenges they must face. Receiving countries
must make a major effort to integrate this student population into their educational sys-
tems, in order to ensure, as far as possible, that equal opportunities are available to all. The
present study analyses the educational trajectory of young people (aged 15–34) in Cat-
alonia (Northeast Spain) and the relationship between the desired and actual level of
education attained. Kaplan–Meier and Cox models (survival analysis) were fitted to assess
the influence of different variables in explaining the time spent in the education system.
After controlling for personal, parental and contextual factors, we found that Moroccan
youth have a higher risk of abandoning their studies, and therefore attain a lower educa-
tional level. However, Latin-American populations are more likely to feel dissatisfied with
the level achieved.
Keywords Education Youth Immigrant population Integration Survival analysis Spain
This work is part of the following research project ‘Geographic mobility and housing: Spain in an inter- national perspective’ (CSO2013-45358-R) and ‘Family challenges at the beginning of the XXIst century: the impact of family individualization on culture, fertility and social welfare’ (CSO2013-46440-P), funded by the Ministry of Economy and Competitiveness.
& Verónica de Miguel-Luken email@example.com
1 Departamento de Derecho del Estado y Sociologı́a, Facultad de Ciencias Económicas y Empresariales, Universidad de Málaga, Campus de El Ejido, 6, 29071 Málaga, Spain
2 Departament de Geografia, Facultat de Filosofia i Lletres, Universitat Autònoma de Barcelona, Barcelona, Spain
Soc Indic Res (2017) 132:733–755 DOI 10.1007/s11205-016-1323-4
During the first years of the 21st century, Spain has become one of the world’s main
recipients of immigrant populations. In 2000, only 2.3 % of the resident population was of
foreign nationality; in 2013 this figure increased to 12 %. The economic crisis that began
in 2008 slowed the pace of arrivals and increased the number of departures, both to return
to a home country or to migrate to a third country, and there was a slight increase in
emigration of Spanish citizens to other countries (Recaño et al. 2015). Nonetheless, the
total foreign population has remained quite stable in recent years, due to ongoing processes
of family reunification as well as increased arrivals from European or various Asian
countries. This increased foreign population, together with the changing characteristics and
the arrival of a growing number of children, adolescents and young adults, challenges the
very notion of external migration held by the resident population and reflected in public
policies. What had been considered a transitory phenomenon becomes a structural reality,
and this leads to a need for immigration and integration policy that transcends the policy
for managing migratory flows (Terrén 2011).
In this context, questions arise about the process of accommodating/integrating this
population and the policies that are needed to respond appropriately to the challenges
posed by this integration effort. One of the basic concerns is the academic and educational
trajectory of young recent arrivals and of the growing number of descendants of recent
immigrants who are appearing in the school population. If we consider education and
training to be essential elements to ensure equality of opportunities and support the abilities
and social mobility of young people, along with a successful transition to adult life, we
need to address several questions. Is there equality for native and immigrant youth? Are
there differentiating factors that help or hinder certain groups? If so, what are they and
what are the best strategies to combat the negative factors? An assessment of the present
living conditions and educational experiences of the youngest immigrants is important to
understanding their process of integration into Spanish society and what the consequences
could be for the future. The events in nearby countries such as France and Great Britain at
the beginning of the 21st century highlight the importance of thinking about this process,
and especially about its weaknesses in order to improve the situation (Alba 2005; Sil-
berman et al. 2007; Dwyer et al. 2008). In Spain, the incorporation of new immigrants into
the educational system is still a very recent phenomenon and the presence of the so-called
‘‘second generation’’1 is very limited and found mostly in the early years of mandatory
education. The task, therefore, is early detection of the needs of this school population and
the gaps in our system that must be addressed to intervene and quickly prevent possible
situations of inequality or conflict.
Several recent reports on youth in Spain have raised alarm, although the problems
identified have long been known about. According to data from the European Union, at the
end of 2014 Spain had the highest unemployment rate (51.4 %) among young people (aged
15–24 years), followed by Greece; the European mean was 21.4 %. The phenomenon of
those categorized as ‘‘not in employment, education or training’’ (NEET) in Spain
exceeded 16 % in 2012, putting it in the group with the highest NEET incidence, along
with Cyprus, Romania, Ireland, Greece, Italy and Bulgaria;2 using 2010 data and the same
1 Although we consider this expression to be incorrect in the sense that the immigrants’ descendants are not immigrants, by definition, we adopt it here because it is widely used in the literature on migrations. 2 European Commission (2015), Youth Unemployment, http://ec.europa.eu/europe2020/pdf/themes/21_ youth_unemployment.pdf. Accessed 16 April 2015.
734 V. de Miguel-Luken, M. Solana-Solana
criteria as Eurostat, Pau Serracant (2014) placed the present study area (Catalonia) in
second place, after Bulgaria.3 The proportion of early leavers from education and training,
28.8 % of men and 20.8 % of women,4 clearly reflects this lack of continuity in formal
preparation and puts Spain at the top of the list for this statistic, and twice the European
Unemployment, economic and job insecurity, and limited school success define the
situation faced by a large part of the youth population of Spain. Although the situation is
better in Catalonia, the problems are basically the same. Little work has been done so far in
the Spanish context, regarding failure or success in school of young immigrants and
descendants of immigrant population, mainly due to the lack of appropriate data. Besides,
the existent previous researches have mainly used a qualitative approach, useless to gen-
eralize results and assess the real extent of the lower performance of students with a foreign
origin. This is the context within which the present article considers the presence of the
immigrant population in the Spanish education system, their aspirations and expectations,
and their educational trajectory. Results will show whether there exists already a risk of
reproduction in the next generations of the etno-stratified labour market and society, and
that immigrants’ descendants (especially those of some specific origins) will be assigned to
the lowest and most stigmatized positions.
2 Theoretical Context: Youth of Immigrant Origin and Educational Attainment
The importance of understanding the integration processes and the reasons for school
success or failure in the young population from immigrant communities has drawn
attention to education and training topics, especially in countries with high immigration
and a long history of managing immigrations. It is not surprising, therefore, that a large
number of the most important studies have been carried out in the United States and
Canada, and somewhat later in Western Europe, especially in countries such as Great
Britain, The Netherlands, or France (Alba et al. 2011). In Spain, and in Catalonia
specifically, new research is beginning to be published that shows some of the most
important educational and workforce entry patterns of this young population of immigrant
In the 2011 census, 11.2 % of the total Spanish population, and 15.1 % of the popu-
lation of Catalonia, were foreigners. Among young people (aged 0–24 years), these rates
were 11.9 % in Spain and 18.9 % in Catalonia. A large part of the foreign population
attending Catalan schools were of Moroccan nationality (Table 1), which corresponds to
the largest immigrant group in Catalonia but also to the first wave of immigration in the
late 1970s from a country other than Europe or Latin America, the ‘‘traditional’’ immigrant
populations in Catalonia. Romanian immigrants should also be highlighted. Relatively
recent arrivals, they have become the second largest immigrant group in Catalonia. The
following Latin American nationalities are also quite well represented, in order of
prevalence: Ecuador, Bolivia, Colombia, Dominican Republic and Honduras. The
Dominican Republic has the longest immigration history; the remainder has arrived in
3 The same author also questions the negative connotations of the concept and the association attributed to their age, proposing more precise indicators. 4 Percentage of the population aged 18–24 with at most a lower secondary education and not in further education or training. Source: Eurostat, Basic figures on the EU (Summer 2013 edition).
Immigrants in the Educational System in Spain: Who Persists? 735
larger numbers since 2000. Finally, there is a growing Asian population, especially from
China to Pakistan. The groups with a longer immigration tradition in Catalonia that persists
to the present have shown the largest population growth in relative terms, despite the
It is not unusual, therefore, that studies to date have focused on youth of Moroccan
origin, not only because of their numbers but also because they have the lowest persistence
rates in completing the mandatory education level, despite the fact that a large proportion
of them were born in Spain and have been incorporated into the Spanish educational
system in a ‘‘normalized’’ manner (Aparicio 2007; Carrasco et al. 2009; Pàmies 2008;
Serra 2010). Another factor to consider is that they are the most stigmatized population in
Spain. Spain’s historical relationship with Morocco (and, by extension, with Muslims) is
long and ridden with conflict. The ‘‘moro’’ (Moor, or Arab) as a social construct in Spanish
imagery persists in many everyday expressions and generates the greatest rejection by the
Spanish population in general (Cea d’Ancona and Vallés Martı́nez 2010).
School ‘‘failure’’ by young Moroccans is often interpreted from a cultural perspective
that emphasizes the lack of interest shown by parents in their children’s studies and school
progress and, in the process, blames them for their ‘‘attitude’’ (Carrasco et al. 2009).
Nonetheless, this lack of parental interest in the process of educating their children is not so
clear in a number of fairly recent studies (Carrasco et al. 2009; Pàmies 2006; Pàmies-
Rovira 2012), especially when a successful migratory project depends in part on providing
better educational and professional opportunities for their children. In the case of suc-
cessful female Moroccans, however, the process has often required extra negotiation, since
it defers the age of marriage and may change family gender roles (Bertrán Tarrés et al.
2014). In a context where discrimination is an everyday experience for Moroccans, it is
obvious that this rejection will have a negative effect on the self-perception and self-esteem
of these young people. This is also true for groups such as Dominicans, Pakistanis, or
Senegambians (see note, Table 1), and in general all immigrants from poor countries who
have non-European/nonwhite phenotypes. Studies in the United States have highlighted
this aspect in Latin American populations and their descendants (Portes and Rivas 2013).
Table 1 Major nationalities of the foreign population attending school in Catalonia, 2013–2014
School population of foreign nationals 2013–2014
Total population of foreign nationals 1 January 2015
Population ranking by nationality
Morocco 52,378 217,955 1
Romania 11,628 99,892 2
China 8169 52,417 3
Ecuador 8025 32,867 8
Bolivia 6330 36,725 6
Pakistan 6175 43,568 5
Senegal-Gambiaa 5372 35,609 7
Colombia 4022 27,045 9
Dominican Republic 3387 19,535 18
Honduras 3011 21,655 11
a These two countries are usually considered as one entity
Source Data provided by the Catalan government, collected by the Department of Education and published by the General Directorate for Immigration (Generalitat de Catalunya)
736 V. de Miguel-Luken, M. Solana-Solana
Nonetheless, some studies indicate that self-esteem can be improved in educational settings
where teachers or professors intervene against discriminatory attitudes and show that they
value a multicultural classroom by working with that aspect of the student profile (Ver-
kuyten and Thijs 2004).
A recent Spanish study by Aparicio and Portes (2014) showed that foreign-born students
have equal aspirations and educational plans than those born in Spain and from native
parents. Other authors (Cebolla Boado and Martı́nez de Lizarrondo 2015) who observed
this same relationship in brute data found that, after controlling for family sociodemo-
graphic characteristics and previous accomplishments, there was greater optimism toward
their studies among pupils of immigrant origins than among the native-born students.
In the United States, Portes and Rivas (2013) also took an approach that emphasizes
cultural patterns to explain the differences they observed between the educational success of
Asian Americans and the higher incidence of academic failure and shorter educational tra-
jectories among students with a Latin-American origin. However, these references to culture
or country of origin must be considered with caution, as shown by pioneering studies that call
into question this direct relationship and point to the importance of contextual factors. For
example, students of Chinese origin achieve low levels of academic success in Spain, in
contrast with the experience—and reputation—of Chinese immigrants in other countries;
furthermore, Yiu (2013) found no direct relationship between academic path and employ-
ment success in this community, given the difficulties of a labor market with high levels of
unemployment and young workers with low qualifications. Therefore, and given this context,
different strategies are needed for success, and training is not necessarily considered the
priority factor in ensuring oneself a good economic future (Portes et al. 2011; Yiu 2013).
Therefore, it is not surprising that the fact of being born in Spain, although it has a
positive impact on a student’s aspirations and expectations, is modest. In fact, years of
residence in Spain have a negative effect that significantly reduces these hopes and plans.
This latter effect is somewhat surprising but can be linked to the contextual elements
referenced above: ‘‘with better knowledge of the destination country comes greater pes-
simism related to the opportunities to achieve progress there’’ (Portes et al. 2011, p. 71). A
more or less optimistic view takes into account both the overall situation and individual,
family and contextual characteristics.
As we consider this game of scale (from the individual to the classroom, from the
teacher to the school and the broader social and cultural context), we cannot forget the
importance of the school as a key educational institution and specifically the role of
teachers in the educational path of boys and girls. It is not only a question of the quantity
and quality of resources. In this sense, the points of reference in the Spanish and specif-
ically in the Catalan context are based on the disjuncture between public schools and the
private or ‘‘concertada’’ (state supported non-public) schools. Students of immigrant origin
are more concentrated in public schools, especially the students from the poorest countries.
In 2012–2013, Spain’s public schools enrolled 84 % of this population, but these students
represented only 13 % of the total school population. In that same year, 64 % of Spanish
nationals attended public schools and 36 % were enrolled in private/non-public schools.5
This is a somewhat simplistic perspective, given the heterogeneity of schools in Spain. A
public school in a small middle-to-high income city in the interior of Catalonia will be very
different from a public school in a neighborhood on the periphery of the city of Barcelona,
with high levels of unemployment and economic insecurity among its population.
5 Data from the Department of Education, Generalitat de Catalunya.
Immigrants in the Educational System in Spain: Who Persists? 737
We have noted the stereotyped view of immigrant parents’ lack of involvement in
following and ensuring the school success of their children. In another direction, which
affects educators’ perception of the students themselves, we must call to mind the research
on ‘‘self-fulfilling prophecy’’ (Rosenthal and Jacobson 1968; Hargreaves et al. 1975),
recently applied by Telles and Ortiz (2008) to the case of the United States.
Recent studies in Catalonia and in Spain as a whole also confirm this idea (Carrasco
et al. 2013; Giliberti 2013a). The first study considered primary and secondary schools with
a high percentage of Moroccan students. Giliberti, on the other hand, analyzed in detail the
practices in different schools with a major presence of Dominican students. In both cases,
the schools were in working-class neighborhoods on the periphery of Barcelona, charac-
terized by a high level of foreign immigration. Not only did teachers mentally classify
students from a success/failure perspective that patently affected the attention paid to the
students in each category, but the school itself was affected: the ‘‘diversity classrooms’’
(aules de diversitat) destined for recent immigrants, and especially those new arrivals with
greater difficult in schools—whether due to previous schooling (both quantity and quality)
or to language difficulties (Hallam and Ireson 2003)—become the ‘‘dummies’’ classes, as
the students themselves call the classrooms (Gibson and Carrasco 2009; Giliberti 2013b).
Although educational authorities do not allow student stratification practices based on
achievement (or, more appropriately, on expectations about a student’s future school
achievement), such practices are common (although hidden) in Catalan schools (Carrasco
et al. 2013; Giliberti 2013b). All this despite the standard implemented in the Catalan
school system in the 1990s that promotes attention to diversity and multi- or intercultur-
alism in the sense of paying attention to the growing diversity in the classrooms and
strengthening content across the curriculum that can prepare students to live in a culturally
diverse society (Garreta Bochaca 2006).
A major connection between the studies carried out in Catalonia and in other contexts is
the importance of persistence in exploring cultural practices from the student’s place of
origin and the processes of multiple identification—such as bilingualism- in academic
achievement by young students. Although some researchers take the position that cultural
assimilation is the best weapon to fight against any kind of discrimination and achieve
positive incorporation in the destination society (Huntington 2004; Alba and Nee 2003),
Portes and Rivas (2013) have refined that simplistic view and introduced a more precise
perspective on a complex process:
A middle ground is that although poorly endowed immigrant families face distinct
barriers to upward mobility, their children can overcome these obstacles through
learning the language and culture of the host society while preserving their home
country language, values, and customs. (Portes and Rivas 2013, p. 219)
The concept of ‘‘segmented assimilation’’ has been used to define these practices that
bridge the here and there, allowing immigrants to build a better future. From this per-
spective, different studies (Carrasco et al. 2009; Pàmies-Rovira 2012) have pointed out the
importance, for example, of the processes of intracultural association (religious, traditional
arts, mutual aid, etc.) and assuming one’s ethnic identity as a young person in the
Moroccan community in Catalonia. It is precisely those most connected to their commu-
nity of origin and participating in ethnic events and associations who attain greater edu-
cational success, compared to those who ‘‘assimilate’’ to the behaviors of Spanish youth.
738 V. de Miguel-Luken, M. Solana-Solana
3 Research Questions and Hypothesis
Despite existing literature that points out differences in educational attainment and
expectations of some foreign groups compared to native populations, there is still a lack of
research that relates the performance deficit in the immigrant student population to other
socio-demographic characteristics and context-specific explanatory variables. The aim of
this paper is to provide new insights about the academic achievement gap in the young
population in Catalonia (Spain) and explore whether differences associated with cultural
and ethnic origin persist after controlling for other variables. Furthermore, we will analyse
the correspondence between aspirations and achievements, in an attempt to assess the
effect of cultural factors on educational patterns.
Transitions to a complete and satisfactory adult life are usually supported by an
appropriate educational path. For a large proportion of the young immigrant population,
however, international mobility could pose an obstacle to continuing their studies, which
may accelerate their transition to the labour market and affect their human capital value.
Our main research questions are whether geographic origin affects the length of time
young people remain in the educational system, after controlling for individual, parental
and contextual variables and, for those who have already quit school, whether the edu-
cation level attained is consistent with their previous expectations. On one hand, this study
analysed the different lengths of time the student population remained in the educational
system and related these patterns to explanatory variables at various levels. On the other
hand, we checked the correspondence between desired and achieved educational attain-
ment for those who were no longer students. Discussing the differences in terms of
immigrant status and specific national origin allows us to assess possible ethnic barriers to
longer trajectories and whether these have an influence on unmet expectations. If this were
the case, it would certainly affect the degree of social integration of the affected popula-
tion, not just in terms of labour fulfillment, but also general well-being.
Our main hypotheses are:
H1 Not all immigrant groups obtain the same results, in terms of years of education, as the native population. The educational system not only does not succeed in balancing the
inequalities of students’ backgrounds and ‘imported’ human capital, it might even help to
perpetuate lower expectations and prejudices toward certain national or ethnic groups, like
H2 The mismatch between time in the educational system and satisfaction with the level of achievement (after controlling for individual and family variables) is more likely to be
associated to a wide range of nationalities, suggesting that migration hastens incorporation into
the labour market for many youths, and may cause an undesired halt in their formal training.
4 Data and Methods
4.1 Data Source
We used the data from the Survey of Youth in Catalonia 2012,6 which gathers information
about transitional processes to adulthood. More specifically, it focuses on the educational,
6 Enquesta a la Joventut de Catalunya de 2012 (EJ12) (http://benestar.gencat.cat/ca/ambits_tematics/ joventut/observatori_catala_de_la_joventut/).
Immigrants in the Educational System in Spain: Who Persists? 739
labour, residential and family transitions experienced by youth. The study population for
the most recent edition of this survey consisted of young adults aged 15–34 living in
Catalonia as of January 1, 2012, according to the official Population Register.7 A final two-
stage probabilistic sample of 3000 individuals, stratified by territory, sex and age, was
drawn. Data were collected through face-to-face interviews. Even if the survey was not
longitudinal in the sense that it did not follow the same sample units through time, it
provides longitudinal information, since it included a wide set of retrospective questions
about diverse topics, education after 15 years in particular (type of studies, dates of
starting-ending, type of educational center, reasons for leaving, etc.). Consequently, it is a
good source to analyse vital trajectories. Notably, the 2012 survey was the first to provide a
representative sample of the young foreign population in Catalonia (Spain).
In accordance with the theoretical approach, the first dependent variable was time in the
educational system beyond age 15 (non-mandatory education). The second dependent
variable was a binary assessment of whether the educational expectations of those who
were no longer in school had been met, entered as a value of 1 if the person stated that he/
she had not achieved the desired academic level, 0 otherwise.
Three dimensions of explanatory and control variables were added to the models. The
first dimension was variables related to the respondent’s sociodemographic characteristics
and human capital. In order to create the variable for geographic origin, we took into
account not only the place of birth of the young person who answered the questionnaire,
but also the parents’ place of birth. If both parents were born in the same foreign country,
we considered this to be the respondent’s country of origin, regardless of the interviewee’s
place of birth. Nonetheless, students born in Spain whose parents were both born outside of
Catalonia constituted a very small proportion of the school population (1 %), due the very
recent and concentrated time period of the major new immigration flows and the fact that
the sample focused on people over 15 (Miguel-Luken 2013). Young people born in Spain
to at least one parent born in Catalonia or elsewhere in Spain were categorized accordingly
(‘‘Catalan’’ or ‘‘rest of Spain’’). If the parents were both from different foreign countries
(extremely few cases in our sample), their child was categorized by his or her place of
birth. Thus, when using the term ‘origin’, we basically mean ‘place of birth’, for the non-
Spanish origins. We grouped the countries of birth/origin for analysis in order to preserve,
as far as possible, a certain internal homogeneity in the territorial units and obtain sufficient
We were also interested in young people born to Spanish parents who moved to Cat-
alonia from elsewhere in Spain; although there are obvious differences in their ease of
integration, compared to other places of origin, the majority language of instruction is
Catalan, which could present an educational challenge. We considered it important to
include this analysis because of the intensive immigration of labour from the rest of Spain
to Catalonia a few decades before the arrival of the larger flows of foreign immigrants. The
integration of those first immigrants was also the subject of much debate and scientific
research that may have some continuity. In addition, the internal migration population
provides a point of comparison with the newer populations with foreign origins.
7 Padrón Continuo de Habitantes (www.ine.es).
740 V. de Miguel-Luken, M. Solana-Solana
To test the effect of the age of incorporation into the host educational system, we used a
categorical (ordinal) variable that distinguishes between those who have always studied in
Catalonia, those who enrolled before primary school (younger than 6 years), those who
arrived during their primary school years (ages 6–11), those who were first enrolled in
mandatory secondary education (ages 12–15) and those who were older when they arrived.
We consider performance in secondary school, asked in the questionnaire in terms of
‘general marks’, a predictor for the length of the educational trajectory. The responses were
coded according to a scale of five categories. Both the categorical and the continuous
treatment of the variable were tested in the models. Since the pattern of influence was
clearly linear and results hardly differed, the continuous variable was preferred. The effect
of parental encouragement (coded using four response categories) was taken into account
when assessing the impact of family influence on the student.
Native language skill is undoubtedly related to the extent of the difficulty in adapting to
the new context. However, since the questionnaire asked only about current use of Spanish,
Catalan or any other language in various situations and did not inquire about proficiency in
any language, this variable was not included in the models.
The cultural dimension, frequently measured by the importance parents give to formal
education and the social capital acquired through the family (Simon 2001; von Below
2007; Ceballo et al. 2014), was assessed with the question about the level of encourage-
ment the respondent received from his/her parents when studying, and the information on
the highest parental academic level and the highest-category parental occupation,
Finally, contextual effects were assessed by the private/public tenure of the schools
attended, self-reports of the encouragement received from teachers, and the general edu-
cational level achieved by the primary school peer group. However, only the variable about
peer level proved to be significant.
4.3 Multivariate Analysis
Since the survey is addressed to a wide range of young people, not all of them had already
terminated their formal training at the time of the interview. On the other hand, given the
chronology of in-flows in Spain, many young immigrants are still studying; adding the
information about their performance is thus relevant in order to evaluate progress in this
aspect of their integration process. Survival analysis allows us to include in the analysis all
those who were still studying, thus gaining added information. Survival analysis is con-
cerned with studying the time between entry to a specific situation (in our case the entry to
non-mandatory education) and a subsequent event (in our case, leaving the educational
system). It was used here to estimate the hazard (or risk) of giving up formal education,
given some explanatory and control variables. A positive regression coefficient for an
explanatory variable means that the hazard is higher; a negative regression coefficient
implies a higher chance of achieving a longer educational trajectory.
The young people who were still studying at the time of the interview (34.4 %) are
considered censored data in the survival analysis (64.8 % had already experienced the
event: finished their studies). The event is defined as completing studies, so value 1
indicates that studies are finished and 0 that the person has not finished yet (censored case).
Survival time (years of study beyond age 15) is measured in months, recoded as years to
show some results more clearly.
Immigrants in the Educational System in Spain: Who Persists? 741
The Kaplan–Meier estimator of the survival function was used to show the cumulative
proportion surviving (studying), depending on the specific geographic origin and sex. This
method is appropriated for small samples (it can be considered a non-parametric method,
Lee and Wang 2003), so it is also convenient for our purposes since all our subsamples
exceed 30 cases. Survival at any time point is calculated as the product of the conditional
probabilities of surviving each previous time interval (Eq. 1):
S tð Þ ¼ Y
i t 1 di
where r is the number of students in school at the beginning of the ith interval, and d the
number of failures (students leaving the educational system) during the same interval.
The Log-rank test was applied to determine regional and gender differences and the age
at which quartiles of each regional population were still in school were calculated. Cox
models were used for the multivariate survival analysis (Box-Steffensmeier and Jones
2004; Miret i Gamundi et al. 2008). Cox’s method is a semi-parametric flexible procedure,
similar to multiple regression analysis, except that the dependent (Y) variable is the hazard
function at a given time. It is considered that it provides robust estimation and, thus, it is a
safe choice model in many situations (assuming proportionality of hazards) (Klein and
Moeschberger 2003), even when the number of observations per variable is small
(Vittinghoff and McCulloch 2006). The model can be summarised in Eq. (2):
h tð Þ ¼ h0 tð Þ exp ðb1x1 þ b2x2 þ þ bpxpÞ; ð2Þ
where h tð Þ is the hazard or risk of leaving school at time t, x1; x2; . . .; xp
explanatory variables and h0 tð Þ is the baseline of underlying hazard function and it cor- responds to the probability of leaving school when all the explanatory variables are zero.
To address the second research question, logistic models were run to determine unmet
educational expectations, with a value of 1 if the respondent would have liked to have
achieved a different level, 0 otherwise. The model can be summarised in Eq. (3a, b):
1 pðxÞ ¼ b0 þ b1x1 þ þ bpxp ð3aÞ
Solving for p(x), this gives:
p xð Þ ¼ e b0þb1x1þþbpxp
1 þ eb0þb1x1þþbpxp ; ð3bÞ
where in our case, p(x) is the probability of unmet educational expectations, conditioned to
the set of explanatory variables x1; x2; . . .; xp
5 Length of the Educational Trajectory by Origin and Sex
As we compared Kaplan–Meier quartiles for the time before leaving the educational
system (in years) and the proportion of the young immigrant population that did not
complete mandatory studies8 (Table 2), we observed very important differences. Almost
8 This is the mandatory secondary education (Enseñanza Secundaria Obligatoria, ESO), normally com- pleted by 16 years old.
742 V. de Miguel-Luken, M. Solana-Solana
half of the young people from the North of Africa (mostly Moroccans)9 for whom studying
was no longer their main activity had not even finished the mandatory education program
(49 %). By comparison, only 10 % of youth with origins in Catalonia, European countries
other than the EU-15 or Latin America (except for Ecuador, Peru and Bolivia) and 4 % of
those born in the EU-15 or in other developed countries had dropped out before completing
their mandatory education. The Kaplan–Meier quartiles (that take into account censored
data) provide a more detailed picture, showing that half of the North-African population,
consistent with the previous result, still remained in the educational system when they were
almost 17 years old. In contrast, the median was age 22–23 for those born in the most
developed European countries or in Catalonia, followed by almost 21 years of age for
respondents with other Spanish origin. This trend was also observed up to the 25 %
percentile: 25 % of the population was still enrolled in the educational system beyond age
25—again, in the case of Spaniards and other EU-15 countries of origin. At the other end
of the spectrum, 75 % of African youth had left the system at 19 years of age.
The relative position of the places of birth according to the cumulative proportion of
students as age increases is shown in Fig. 1 (censored data are not marked in the figure to
improve legibility). In general, youth with a Catalonian origin achieved better perfor-
mance, with fewer students leaving formal education before the age of 21–22 years
(approximately). After that point, which could coincide with the end of university studies,
longer trajectories were also observed in those with origins in Europe-15 (but not the rest
of Europe), other developed countries, and the rest of Spain.
At the bottom of the scale were the African territories. The pattern is very similar for
Northern African countries and the rest of the continent up to age 19–20, when about 20 %
of this population was still in the education system. Beyond that point, higher proportions
of shorter trajectories corresponded to North-African origins.
Table 2 Percentage of young people who completed mandatory studies and Kaplan–Meier survival quartiles for the age when education ended, by origin
n % of mandatory studies unfinished (base: non-students)
Age at which different proportions of students remain in the educational system
75 % Median 25 %
Europe 15, EEAa, Switzerland ? rest of developed countries
39 3.8 18.33 22.58 25.08
Rest of Europe 117 11.5 17.25 18.75 20.92
Ecuador, Peru and Bolivia 70 14.0 17.25 19.25 21.83
Rest of Latin America 155 10.1 17.67 19.92 23.08
North Africa 182 49.4 15.33 16.83 19.08
Rest of Africa 56 37.0 15.58 17.17 19.08
Catalonia 1939 10.2 18.83 22.08 25.25
Rest of Spain 429 16.5 17.75 20.67 24.58
a European Economic Area
Source own elaboration from the EJC 2011 microdata
9 In the sample, 96.6 % of the young people born in the Maghreb, were born in Morocco.
Immigrants in the Educational System in Spain: Who Persists? 743
In between these extremes, we find the rest of the Latin-American countries, followed
by other European origins. At this stage, without taking into account any other variable that
could help to explain these differences, we observed a sort of hierarchy in the patterns:
Spanish and EU-15 origins showed longer educational paths, followed by Spanish-
speaking Latin American origins and, finally, other non-Spanish speaking origins.
To assess withdrawal, only those who quit school in the midst of an educational level
(primary grades, middle school, etc.) were asked the reason they dropped out (i.e., why
they quit a level they had begun but did not complete). Therefore, the sample for each
origin was dramatically reduced, which precluded drawing conclusions. However, it is
interesting to point out that a large share (about 19 %) of the foreign-born young popu-
lation who interrupted their studies gave ‘other’’ as their reason (not recoded in the data
file), suggesting that their own migration movement could be the reason. Of the respon-
dents who had immigrated on their own, as a personal project, 47 % declared that they
years studying after 15
catalonia rest of Spain rest of Africa North-Africa rest Latinamerica Ecuador, Peru & Bolivia rest Europe
Europe 15, EEE, Switz + rest developed countries
Fig. 1 Cumulative proportion of students ages 15–34 by years studying after age 15 and place of birth. Significant log-rank test. Source own elaboration from the EJC 2011 microdata
Table 3 Kaplan–Meier survival median for the age when educa- tion ended, by origin and sex
a European Economic Area
Source own elaboration from the EJC 2011 microdata
n Median n Median
Rest of Europe 54 18.25 63 19.25
Ecuador, Peru and Bolivia 30 18.67 40 20.83
Rest of Latin America 69 19.17 86 20.17
North Africa 83 17.50 97 16.33
Catalonia 981 21.67 953 22.42
Rest of Spain 217 19.83 209 21.00
744 V. de Miguel-Luken, M. Solana-Solana
years studying after 15
catalonia rest of Spain North-Africa rest Latinamerica Ecuador, Peru & Bolivia rest Europe
years studying after 15 20151050
Fig. 2 Cumulative proportion of students ages 15–34 by years studying after age 15 and place of birth and sex. Significant log-rank tests. Source own elaboration from the EJC 2011 microdata
Immigrants in the Educational System in Spain: Who Persists? 745
abandoned studies for economic reasons or because of a job opportunity; this was the main
cause stated by 35 % of those who immigrated with their family members and 20 % of the
Spanish-born (Miguel-Luken 2013).
Table 3 shows the survival medians, as provided by the Kaplan–Meier survival pro-
cedure, by sex and origin (some of the origins have been omitted because of their smaller
sample sizes). Females perform better than males for all origins except for the case of
North African countries. However, the observed rank by region of origin is similar for both
sexes. At age 17.5 already half of the male population born in North Africa has already
quitted school, age that decreases to 16.3 for females from the same origin. These are
followed by the youth immigrated from Eastern European countries (18.25 for men and
19.25 for women) and, then, Latin-American countries. At the other end, it is found that
50 % of the Catalonian-origin females still study at age 22.4, while 50 % of Catalonian-
origin males remain in school at 21.7.
The whole pattern through time can be observed at Fig. 2. Curves do not differ much
according to their relative position with regards to the others. However, the most
remarkable differences by sex are for the Ecuadorians, Bolivians and Peruvians, for whom
there is a sharper and more rapid decrease during the first years after 15 for the boys
(a) whilst the abandoning is more gradual for the girls (b). The opposite relation is found
for North-Africans. More than 20 % of the North African females do not study anymore at
age 15, and the cumulative proportion still at school at any age is always under the one for
6 Differential Effects on the Length of the Educational Path
We built models for three levels of effects on educational attainment (measured as time in
formal education as the main activity) considered in the present study: individual (socio-
demographic profile and human capital), parental (culture and social capital) and con-
textual (influence of teachers and peers). The baseline model included only sex and geo-
graphical origin (Table 4).
6.1 Individual Effects
When no explanatory variables other than sex and origin were considered (model 1,
Table 4), the risk of leaving school was higher for students of any geographical origin
other than Catalonia. Even the young people with other Spanish origins dropped out earlier
than those with no immigrant origin at all. However, the pattern is not equal for all national
origins, as illustrated by the differences with regards to the baseline category. North-
African young immigrants are more likely to experience shorter educational trajectories,
followed by the rest of the African countries and the rest of Europe (not in Europe-15).
The literature shows that young women have better educational outcomes than their
male counterparts, in general, in contemporary European or Western societies (Buchmann
et al. 2008; Miret i Gamundi et al. 2008; Ministerio de Educación 2011). However, this
effect disappears with the addition to the models of other explanatory variables (Table 4,
Even taking into account the control variables (age of entry into the Spanish educational
system and secondary school marks), many geographical differences still persisted (second
model, Table 4). In fact, the three national origins that occupied the worst positions in the
746 V. de Miguel-Luken, M. Solana-Solana
m o d el
n g th
ed u ca
ti o n al
o u n g
p o p u la
ti o n .
al o n ia
, 2 0 1 1
o d el
S ex M
3 6 *
E u ro
1 5 ,
.1 4 2
.0 7 4
.5 9 4
ad o r,
.2 6 1
.3 1 6
.5 8 0
.1 4 7
.2 3 1
.1 2 8
.5 2 6
.2 4 8
.4 9 3
.1 8 1
.0 4 4
m p le
3 2 *
1 6 *
4 4 *
3 7 *
– 1 5
1 6 ?
5 4 *
6 8 *
co n d ar
h o o l
0 2 *
t p ar
ti o n al
le v el
o n d
4 7 *
8 1 *
Immigrants in the Educational System in Spain: Who Persists? 747
n u ed
o d el
t p ar
l o cc
u p at
g o ry
d g en
0 6 *
.2 2 9
L o w
ry o cc
u p at
io n s
.1 5 3
0 3 *
o n d
8 4 /1
lo g -l
8 8 4
9 4 5
* p \
* p \
* p \
S o u rc e
ab o ra
748 V. de Miguel-Luken, M. Solana-Solana
first model continued at the same rank. This is an interesting finding because it would seem
that the energy and effort required to adapt to a new context and, more precisely, to a new
academic environment, would be highly relevant to explaining the length of the educa-
tional trajectories after immigration, thus compensating for the influence of national origin.
However, we found that the coefficients for these birthplaces remained significant. Fur-
thermore, the disadvantage remained for the descendants of internal Spanish immigrants.
This leads us to argue that modes of incorporation—adapting the Portes and Rumbaut
(2001) terminology to our educational context—are probably not the same for all ethnic
groups. This finding could be explained to some extent by the effect of native language
skills, since young Latin-American immigrants who speak Spanish (but very rarely have
any initial Catalan-language proficiency) did not have significant differences, compared to
the local population; their language skills could be an advantage, compared to African
immigrants. Nonetheless, that would not help to explain the better persistence of European-
15 immigrants, for whom other reasons associated to their higher socioeconomic back-
ground may lie behind the results.
The age of incorporation into the school system in the destination country had the
expected impact. The earlier the immigrant entered the educational system in the new
country, the longer the educational trajectory, compared to those who completed all of their
studies elsewhere (normally in the country of origin). For an important portion of this
young population, this finding suggests that emigration could mean a disruption in their
academic careers, or that many foreign immigrants are likely to arrive with a lower formal
training level than the Spaniards.
Finally, immigrants who started their studies in Spain after age 16 have a slightly lower
risk of dropping out than those who started between ages 6 and 15. This group includes
people for whom studies were the main reason for immigrating (for instance, post-doctoral
students), and their presence in the sample would obviously have an effect that offsets the
6.2 Parental Effects
In model 3 (Table 4), we tested the effect of inherited family social capital through the
highest academic level achieved by father or mother, together with the highest parental
socioeconomic level. These explanatory variables allowed us to test the hypothesis that
social reproduction accounts for a great deal of academic success, measured here in
terms of years in the educational system. As expected, the lower the parents’ academic
attainment, the higher the risk of abandoning the studies. Similarly, young people
coming from a family with a medium-rank occupational category performed significantly
worse than those whose mother or father worked as a general director or as a manager in
We also wanted to test the effect of parental expectations and encouragement (cultural
effects), for which we included a variable on the extent to which the respondent thinks his/
her parents pushed him/her to continue their studies. The higher this parental support, the
lower the risk of leaving school.
So far, the new explanatory variables followed the expected pattern. It is important to
highlight that most of the other influences still remain after controlling for family effects.
Although generally attenuated, the estimates for age of incorporation into the Spanish
educational system and the grades achieved in compulsory secondary education pointed in
the same direction as in the previous models. With regards to geographic origin, only the
Immigrants in the Educational System in Spain: Who Persists? 749
coefficient for ‘North-Africa’ remains significant, so there is still some evidence of a
significantly higher likelihood for this group of a shorter stay in the formal educational
6.3 Contextual Effects
Finally, we tried to capture the impact of the context on the individual educational tra-
jectory. There is some information in the questionnaire that can serve as proxies to test this
contextual influence, but most proved to be non-significant in the models. For instance,
mainly attending private vs public schools did not have a significant effect; neither did the
encouragement of teachers to continue studying. However, including information about the
level of education achieved by other primary school classmates to assess the question of
the expectations created by the peer group showed that young people whose classmates
mostly graduated from university had a lower risk of early departure from the system. In
other words, they are the respondents more likely to follow longer trajectories, as their
The most important finding, however, was the confirmation that after controlling for all
these variables, the North-African population was still significantly more likely to have
shorter academic trajectories, which is very consistent with the results of the cited studies
carried out from a more qualitative perspective. In summary, there was some evidence that
hidden aspects are operating that hinder the educational possibilities of this collective, in
large part composed of Moroccan youth. Social expectations, in the broad sense, may
partially explain this result, due to the manifest prejudices and stereotypes of the general
population in Spain toward this group (Cebolla and Requena 2009).
7 Fulfilled Expectations
Finally, we analysed whether national origins were positioned similarly in the mismatch
between the achieved and desired level of education among our respondents. This
analysis was limited to the respondents who were no longer enrolled in a formal edu-
We followed the same steps as for the previous models, although the significant vari-
ables were considerably reduced (for instance, contextual variables were no longer rele-
vant). If only sex and place of birth were taken into account (model 1, Table 5), Latin
Americans in general and North Africans were significantly more likely to be dissatisfied
with their academic achievement. This is more notorious in the case of Ecuadorians,
Peruvians and Bolivians and consistent with the previous findings regarding time spent in
the educational system. The only exception was found for Asians, with a lower probability
of being dissatisfied than Catalan young people (baseline category).
Nonetheless, as age of incorporation into the Spanish educational system is added to the
model, the significance of the estimators persists for the previous groups and the associated
coefficients grow. It is also interesting that children who entered Spanish schools before the
age of 6 fulfilled their own expectations, but the probability of unmet expectations was
higher for those immigrants who arrived between ages 6 and 11, and slightly higher for
those who arrived in adolescence (12–15 years). Older students who immigrated after age
15 specifically for a university education are included in the category ‘16?’, which helps
750 V. de Miguel-Luken, M. Solana-Solana
to explain why the trend regarding age of entering the education system in Catalonia was
not completely linear.
The influence of parents’ socioeconomic and academic level was considered in model 3.
There was some evidence that sons and daughters of parents with a secondary education
were more likely to have expressed discomfort about their academic performance, whilst
Table 5 Logistic models for the match between achieved and desired educational level
Model 1: basic characteristics
Model 2: human capital
Model 3: parental effects
B exp(B) B exp(B) B exp(B)
Female 0.154 1.167 0.162 1.175 0.167 1.182
Place of birth
Europe-15, EEA, Switzerland ? other developed countries
-0.641 0.527 0.097 1.102 0.070 1.073
Other European countries -0.255 0.775 0.594 1.812 0.502 1.652
Ecuador, Peru and Bolivia 2.279** 9.769 3.143** 23.169 3.192** 24.334
Other Latin American countries 1.042*** 2.834 1.933** 6.910 1.921** 6.824
North Africa 0.669** 1.953 1.373* 3.949 1.491** 4.443
Other parts of Africa 0.528 1.696 1.390* 4.015 1.442* 4.229
Spain (outside Catalonia) 0.137 1.147 0.145 1.156 0.163 1.176
Age of incorporation into the Spanish educational system
Completed all studies abroad
0 1.016 2.761 0.956 2.600
\6 0.848 2.336 0.898 2.454 6–11 0.996* 2.706 0.852 2.345
12-15 1.054* 2.870 0.974* 2.649
16? 0.245 1.278 0.260 1.297
Highest parental education level
Compulsory or less
Secondary 0.294** 1.342
Postsecondary/university -0.008 0.992
Highest parental occupation category
Managers and general directors
Professionals and technicians -0.232 0.793
Medium-rank occupations 0.040 1.040
Low-rank/elementary occupations -0.242 0.785
Other -0.642** 0.526
Constant 0.731*** 2.076 -0.288 0.750 -0.238 0.788
n 1803 1803 1787
-2 log-likelihood 2087.00 2078.50 2042.05
* p\ 0.1; ** p\ 0.05; *** p\ 0.001 Source own elaboration from the EJC 2011 microdata
Immigrants in the Educational System in Spain: Who Persists? 751
the opposite is observed for the offspring of parents in the occupational category of ‘other’
(mainly retirees or unemployed parents).
Once the effect of the family background was tested, the importance of entry between
age 6 and 11 decreased (no longer significant), but the positions according to the place of
origin did not change. In summary, young people born in Ecuador, Peru and Bolivia who
arrived as teen-agers and whose parents had completed secondary education were more
prone to state that they had not reached their desired degree of education. These were
followed by the rest of Latin Americans and Africans.
Worse performance by immigrants or their descendants at school, compared to their native
peers, has been reported for various geographical settings and immigrant populations. To
some extent, our results are quite consistent with these previous findings (although now
based on a representative sample and quantitative methods), including those specifically
focused on Spain: immigrants get poorer results (measured here by length of the educa-
tional trajectory), especially those from North-African countries (where Morocco is, by far,
the most represented origin).10
However, we provide some evidence that this disadvantage persists after controlling for
important individual, parental, and contextual variables. It is not just that North-African
young people in Catalonia follow shorter educational paths; the important finding is that
this is true regardless of their age at entry into the Spanish educational system, their
previous performance at school, the role and background of their parents, the role of
teachers, and the achievement of their former classmates. Our most important contribution
is showing that their lower achievement is not completely related to the socio-economical
background of the family or their longer or shorter experience in Spain. This leads us to
continue looking for an explanation of this gap and urges us to act. Appropriate inter-
ventions are needed to eliminate differences that could rest on some sort of stereotype that
makes it much more difficult for this population of North-African origin to successfully
achieve full integration.
Regarding the first research question, further study is needed to clarify the effect of
peers and society on the educational attainment of the immigrant population, particularly
those who persist the least in the formal educational system. Some authors alert us to the
over-emphasized impact on student expectations of contextual factors such as the pro-
portion of immigrant students in schools (Cebolla Boado and Martı́nez de Lizarrondo
2015). Nonetheless, social expectations associated to certain ethnic groups may still have a
subtle and relevant influence on young students. The boundaries of the contextual influence
on the individual are not the school walls, but the neighbourhood, the friends and
acquaintances, the media, and society in general. The way others perceive the young
immigrant’s possibilities of success or failure can generate a self-fulfilling prophecy in
students who are more vulnerable to stereotypes. In this sense, it could be very useful to
study the individual together with his or her friendship networks. Constrained by the use of
secondary data on educational levels attained by the study population’s primary school
10 It has to be pointed out that the lack of significance observed for other African origins could be a consequence of the considerably smaller simple size of this category. Future research should focus more extensively on youth from other origins (also Asian countries) to be able to confirm the worst position already observed for Maghrebians.
752 V. de Miguel-Luken, M. Solana-Solana
classmates, our approach to this question could be considered naı̈ve, but a next step could
be to apply social network analysis to longitudinal data (Lomi et al. 2011) in an analysis of
these observed differences in performance.
Nonetheless, the correspondence between years of study and satisfaction with the
obtained academic level according to place of birth is not obvious. Although young
Maghrebians spend the shortest time in the educational system, there is a higher likelihood
that Latin Americans (mostly from Ecuador, Peru and Bolivia) will declare that they would
have liked to reach a higher academic level, particularly if they arrived to Spain as
adolescents. A good structure of second opportunities could help to attract these immi-
grants into the educational system again so they can continue with their studies when
In brief, the immigrant population in general, but mainly the young North-African
population, seems to face higher constraints and difficulties to achieve the same results as
immigrants from other origins or native youth in terms of years of formal education. After
controlling by sex and social capital (acquired through their parents), the results show that
this population, together with the young Latin American immigrants, are not satisfied with
the level achieved. These findings pose new challenges to educators in multicultural
A successful transition to adulthood contributes to a more integrated society. A more
egalitarian educational system that can help each individual achieve the level of educa-
tional attainment he or she desires will result in a better prepared and more self-actualized
Acknowledgments We would like to thank the Secretaria per a la Joventut, of the Generalitat de Catalunya, for providing us with the microdata file for the analysis and all the information needed about the survey. We would also like to thank the reviewers for their valuable comments, which have improved the preliminary version of this paper.
Alba, R. (2005). Bright vs. blurred boundaries: Second generation assimilation and exclusion in France, Germany, and the United States. Ethnic and Racial Studies, 28(1), 20–49. doi:10.1080/ 0141987042000280003.
Alba, R., & Nee, V. (2003). Remaking the American mainstream: Assimilation and contemporary immi- gration. Cambridge: Harvard University Press.
Alba, R., Sloan, J., & Sperling, J. (2011). The integration imperative: The children of low-status immigrants in the schools of wealthy societies. Annual Review of Sociology, 37, 395–415.
Aparicio, R. (2007). The integration of the second and 1.5 generations of Moroccan, Dominican and Peruvian origin in Madrid and Barcelona. Journal of Ethnic and Migration Studies, 33(7), 1169–1193. doi:10.1080/13691830701541713.
Aparicio, R., & Portes, A. (2014). Crecer en España. La integración de los hijos de los inmigrantes. Barcelona: Obra Social ‘‘La Caixa’’.
Bertrán Tarrés, M., Ponferrada-Arteaga, M., & Pàmies Rovira, J. (2014). Gender, family negotiations and academic success of young Moroccan women in Spain. Race, Ethnicity and Education. doi:10.1080/ 13613324.2014.946486.
Box-Steffensmeier, J. M., & Jones, B. S. (2004). Event history modeling. A guide for social scientists. New York: Cambridge University Press. doi:10.1017/CBO9780511790874.
Buchmann, C., DiPrete, T. A., & McDaniel, A. (2008). Gender inequalities in education. Annual Review of Sociology, 34, 319–337.
Carrasco, S., Pàmies, J., & Bertran, M. (2009). Familias inmigrantes y escuela: Desencuentros, estrategias y capital social. Revista Complutense de Educación, 20(1), 55–78.
Immigrants in the Educational System in Spain: Who Persists? 753
Carrasco, S., Pàmies, J., Ponferrada, M., Ballestı́n, B., & Bertran, M. (2013). Segregación escolar e inmi- gración en Cataluña: Aproximaciones etnográficas. Emigra Working Papers, 126, 1–26.
Cea d’Ancona, M. A., & Vallés Martı́nez, M. (2010). Xenofobias y xenofilias en clave biográfica. Madrid: Siglo XXI.
Ceballo, R., Maurizi, L. K., Suarez, G. A., & Aretakis, M. T. (2014). Gift and sacrifice: Parental involvement in Latino adolescents’ education. Cultural Diversity and Ethnic Minority Psychology, 20(1), 116–127.
Cebolla Boado, H., & Martı́nez de Lizarrondo, A. (2015). Las expectativas educativas de la población inmigrante en Navarra. >Optimismo inmigrante o efectos de escuela? Revista Internacional de Soci- ologı́a, 73(1), 1–13. doi:10.3989/ris.2013.02.22.
Cebolla, H., & Requena, M. (2009). Los inmigrantes marroquı́es en España. In David Reher & Miguel Requena (Eds.), Las múltiples caras de la emigración en España (pp. 251–287). Madrid: Alianza.
Dwyer, C., Bindi, S., & Sanghera, G. (2008). ‘From cricket lover to terror suspect’—Challenging repre- sentations of young British Muslim men. Gender, Place and Culture: A Journal of Feminist Geog- raphy, 15(2), 117–136. doi:10.1080/09663690701863208.
Garreta Bochaca, J. (2006). Ethnic minorities and the Spanish and Catalan educational systems: From exclusion to intercultural education. International Journal of Intercultural Relations, 30(2), 261–279. doi:10.1016/j.ijintrel.2005.11.006.
Gibson, M. A., & Carrasco, S. (2009). The education of immigrant youth: Some lessons from the U.S. and Spain. Theory into Practice, 48(4), 249–257. doi:10.1080/00405840903188118.
Giliberti, L. (2013a) La condición inmigrante y la negritud en la experiencia escolar de la juventud dominicana: estigmas y formas de agencia. Una etnografı́a transnacional entre la periferia de Barcelona y Santo Domingo. Ph.D. Thesis: Universidad de Lleida.
Giliberti, L. (2013b). Escuela y reproducción social: las prácticas ocultas en los sistemas educativos español y dominicano. Mondi Migranti. Revista di studi e ricerche sulle migrazioni internazionali, 2, 221–238.
Hallam, S., & Ireson, J. (2003). Secondary school teachers’ attitudes towards and beliefs about ability grouping. British Journal of Education Psychology, 73(3), 343–356. doi:10.1348/ 000709903322275876.
Hargreaves, D. H., Hester, S., & Mellor, F. J. (1975). Deviance in Classrooms. London: Routledge & Kegan Paul.
Huntington, S. (2004). The hispanic challenge. Foreign Policy, 14, 30–45. doi:10.2307/4147547. Klein, J. P., & Moeschberger, M. L. (2003). Survival analysis: Techniques for censored and truncated data
(2nd ed.). New York: Springer. Lee, E. T., & Wang, J. W. (2003). Statistical methods for survival data analysis (3rd ed.). Belmont, CA:
Lifetime Learning Publications. Lomi, A., Snijders, T. A. B., Steglich, C. E. G., & Torló, V. J. (2011). Why are some more peer than others?
Evidence from a longitudinal study of social networks and individual academic performance. Social Science Research, 40, 1506–1520. doi:10.1016/j.ssresearch.2011.06.010.
Miguel-Luken, V. de (2013). Població d’origen estranger. Transicions accelerades en un context de vul- nerabilitat. Enquesta a la joventut de Catalunya 2012 (Vol. 2, pp. 219–266). Barcelona: Secretaria per a la Joventut, Generalitat de Catalunya.
Ministerio de Educación (2011). Panorama de la educación. Indicadores de la OCDE 2011. Informe español. Madrid.
Miret i Gamundi, P., Salvadó i Nayach, A., Serracant i Melendres, P. & Soler i Martı́, R. (2008). Enquesta a la joventut de Catalunya 2007. Una anàlisi de les transicions educatives, laborals, domiciliars i familiars. Barcelona: Observatori Català de la Joventut IV. Collecció: Estudis, 24.
Pàmies, J. (2006). Dinámicas escolares y comunitarias de los hijos e hijas de familias inmigradas mar- roquı́es de la Yebala en la periferia de Barcelona. Ph.D. Thesis, Universitat Autònoma de Barcelona.
Pàmies, J. (2008). Identitat, integració i escola. Joves d’origen marroquı́ a la perifèria de Barcelona. Barcelona: Secretaria de Joventut. Generalitat de Catalunya.
Pàmies-Rovira, J. (2012). Moroccan Immigrants at a Secondary School in Catalonia. In Spinthourakis, J. A., J. Lalor &W. Berg (Eds.), Cultural diversity in the classroom. A European comparison. VS Research, pp. 79–94.
Portes, A., Aparicio, R., Haller, W., & Vickstrom, E. (2011). Progresar en Madrid: aspiraciones y expec- tativas de la segunda generación en España. Revista Española de Investigaciones Sociológicas, 134, 55–86. doi:10.5477/cis/reis.134.55.
Portes, A., & Rivas, A. (2013). The adaptation of migrant children. The Future of Children, 21(1), 219–246. doi:10.1353/foc.2011.0004.
Portes, A., & Rumbaut, R. G. (2001). Legacies: The story of the immigrant second generation. Berkeley, CA: University of California Press.
754 V. de Miguel-Luken, M. Solana-Solana
Recaño, J., Roig, M., & de Migue, V. (2015). Spain: A new gravity centre for Latin American migration. In A. Domingo i Valls, et al. (Eds.), Demographic analysis of Latin American immigrants in Spain. Cham: Springer.
Rosenthal, R., & Jacobson, L. (1968). Pygmalion in the classroom. New York, NY: Holt. Serra, C. (2010). Polı́tica a ciegas. Déficit de atención en el seguimiento de las trayectorias académicas del
alumnado inmigrante en el paso de los estudios obligatorios a los postobligatorios. In Grupo Inter- disciplinario de Investigador@s Migrantes (coord.) (Ed.), Familias, niños, niñas y jóvenes migrantes. Rompiendo estereotipos (pp. 139–147) Madrid: IEPALA editorial.
Serracant, P. (2014). A brute indicator for a NEET case: Genesis and evolution of a problematic concept and results from an alternative indicator. Social Indicators Research, 117(2), 401–419. doi:10.1007/ s11205-013-0352-5.
Silberman, R., Alba, R., & Fournier, I. (2007). Segmented assimilation in France? Discrimination in the labour market against the second generation. Ethnic and Racial Studies, 30(1), 1–27. doi:10.1080/ 01419870601006488.
Simon, B. S. (2001). Family involvement in high school: Predictors and effects. NASSP Bulletin, 85(627), 8–19.
Telles, E. E., & Ortiz, V. (2008). Generations of exclusion (Mexican-Americans, Assimilation, and Race). New York: Russell Sage Foundation.
Terrén, E. (2011). Identidades desterritorializadas. El sentimiento de pertenencia nacional entre los ado- lescentes de familias inmigradas. Papers. Revista de Sociologia, 96(1), 97–116.
Verkuyten, M., & Thijs, J. (2004). Global and ethnic self-esteem in school context: Minority and majority groups in the Netherlands. Social Indicators Research, 67, 253–281. doi:10.1023/B:SOCI.0000032339. 86520.5f.
Vittinghoff, E., & McCulloch, C. E. (2006). Relaxing the rule of ten events per variable in logistic and Cox regression. American Journal of Epidemiology, 165(6), 710–718. doi:10.1093/aje/kwk052.
von Below, S. (2007). What are the chances of young Turks and Italians for equal education and employment in Germany? The role of objective and subjective indicators. Social Indicators Research, 82, 209–231. doi:10.1007/s11205-006-9038-6.
Yiu, J. (2013). Calibrated ambitions: Low educational ambition as a form of strategic adaptation among Chinese youth in Spain. International Migration Review, 47(3), 573–611. doi:10.1111/imre.12037.
Immigrants in the Educational System in Spain: Who Persists? 755