Content uploaded by David Glauser
Author content
All content in this area was uploaded by David Glauser on Jul 27, 2016
Content may be subject to copyright.
Content uploaded by David Glauser
Author content
All content in this area was uploaded by David Glauser on Jun 16, 2016
Content may be subject to copyright.
VET or general education? Eects
ofregional opportunity structures
oneducational attainment inGerman‑speaking
Switzerland
David Glauser1* and Rolf Becker1,2
Background
e transition from school-to-work and, consequently, the acquisition of vocational or
general qualifications, is an important step in the life of adolescents with respect to their
subsequent educational opportunities and career prospects over the life course. Due to
Abstract
Background: Regional opportunity structures have received little attention in
research on educational aspirations and attainment. This papers aim is to address
whether regional opportunity structures affect educational attainment at the transition
from lower to upper secondary education net of institutional and individual effects.
Related with this issue we also ask how opportunity structures influence pupils’ educa-
tional attainment.
Methods: We use data of the DAB panel study that contains information on pupils’
educational aspirations as well as their actual educational attainment immediately
after as well as 15 months after leaving compulsory schooling (N = 2.192). In order to
account for regional variation of the opportunity structures we construct one scal-
ing variable based on principal component factor analysis and merge these with DAB
data. In the scaling variable, various aspects of the regional labor market structures and
conditions, as well as the diversity and extent of the supply of educational alternatives,
are considered. Educational outcomes are analyzed using multinomial logistic regression
models.
Results: Our results indicate, first, that the more extensive regional opportunity struc-
tures are, the higher the probability of pupils attending general education programs.
In contrast, restricted opportunity structures increase pupils’ probability of attending
VET. Second, regional opportunity structures are perceived, subjectively evaluated
and organize a pupil’s set of feasible actions and educational attainment: on the one
hand regional opportunity structures are correlated with pupils’ realistic educational
aspirations. Once we control for educational aspirations the direct effects of opportu-
nity structures on educational attainment after leaving compulsory schooling can be
partially explained.
Keywords: School-to-work, Upper secondary education, Opportunity structures,
Educational aspirations, Inequality of educational opportunities
Open Access
© 2016 The Author(s). This article is distributed under the terms of the Creative Commons Attribution 4.0 International License
(http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium,
provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and
indicate if changes were made.
RESEARCH
Glauser and Becker Empirical Res Voc Ed Train (2016) 8:8
DOI 10.1186/s40461‑016‑0033‑0
*Correspondence:
david.glauser@edu.unibe.ch
1 Institute of Educational
Science, University of Bern,
Fabrikstrasse 8, 3012 Bern,
Switzerland
Full list of author information
is available at the end of the
article
Page 2 of 25
Glauser and Becker Empirical Res Voc Ed Train (2016) 8:8
the long lasting consequences, school-to-work related inequality of educational oppor-
tunity (IEO) has been a key subject of empirical educational research for years. On the
one hand, educational opportunities depend on and are provided by the organization
of a specific educational system, i.e., the degree of stratification, standardization, voca-
tional specificity, and permeability. On the other hand, IEO is partially explained by
the differential societal orders and contexts in which pupils are embedded: social ori-
gins, peer-groups, school classes, etc. Also significant—although often neglected—are
regional opportunity structures that may encourage pupils to take up vocational educa-
tion and training (VET) rather than continuing with general education. ese include:
the regional supply of VET and general education programs, local labor market condi-
tions, the size of the school-leaver cohort, and competition among pupils for a small
number of apprenticeships or general education programs.
erefore, we focus on the relevance of regional opportunity structures to the actual
decision as to whether to begin VET or general education programs at the transition
from compulsory to upper secondary education in German-speaking Switzerland. In
order to explain disparities at this particular educational transition, we not only con-
trol for individual characteristics (social origins, GPA, educational aspirations, gender)
and institutional constraints (type of school attended at lower secondary level) but also
use macro level data to consider the regional opportunity structure. Even if the available
administrative data are far from perfect, we address the question of whether and in what
direction regional opportunity structures—in addition to constraints at the individual
and institutional levels—affect pupils in their decision to attend either VET or general
education programs. In addition and related to this issue, we focus on how opportunity
structures influence pupils’ educational attainment.
is paper is structured as follows: In the second section, we describe the education
and training alternatives available under the Swiss education system. In section three, we
describe our theoretical model, which accounts for individual, institutional, and regional
constraints and opportunities in order to explain the educational attainment of youths
and to establish the micro-macro-link. Subsequently, we explicate our data, the opera-
tionalization of the variables used in the multivariate analyses, and the statistical pro-
cedure. We then discuss the analytical results concerning the educational situation of
youths both immediately after and 15months after leaving compulsory education. In
our conclusion, we draw attention to the limitations of our results and to potential areas
for further research.
Educational alternatives aftercompulsory schooling
Upper secondary education in Switzerland begins after 9th grade (for a schematic
representation of the Swiss education system, see Fig.2 in the Appendix). e major-
ity—approximately 70%—of the school leaver cohorts then begin either dual appren-
ticeships or school-based vocational education and training programs in which they
acquire occupation-specific qualifications. Depending on their school performance and
the school track they attended at lower secondary level, pupils can choose from approx-
imately 240 different VET programs. Apprenticeships or school-based VET programs
can be completed either within 2 (Federal VET Certificate, EBA) or 3–4years (Federal
VET Diploma, EFZ; Federal Vocational Baccalaureate Certificate, FVB). An increasing
Page 3 of 25
Glauser and Becker Empirical Res Voc Ed Train (2016) 8:8
proportion of pupils now takes the FVB, which places greater emphasis on general edu-
cation. e FVB can be obtained during or upon completion of a Federal VET Diploma.
General full time education is provided at baccalaureate schools (Gymnasiale Matu-
ritätsschulen) and at specialized schools (Fachmittelschulen, FMS), although the cur-
ricula of the latter are focused more on providing occupation-specific skills. Graduates
of the FVB or FMS are eligible for direct admission to universities of applied sciences,
whereas students with matriculation certificates are entitled to enroll at universities.
e distribution of upper secondary qualifications attained in German-speaking Swit-
zerland in the year 2013 was as follows: EFZ/EBA: ♂: 71.8%, ♀: 63.1%; FVB: ♂: 13.5%, ♀:
12.5%; baccalaureate and specialized schools: ♂: 14.7%, ♀: 24.4% (BFS 2014, owncalcu-
lations). Pupils who—for whatever reason—are unable to immediately commence a cer-
tifying upper secondary VET or general education program after compulsory schooling
are provided with opportunities to begin different types of bridge year course, e.g., 10th
school year, au-pair, pre-apprenticeships, preparatory courses for particular education
programs, etc. ese non-mandatory bridge year courses are accessed by approximately
20% of the school leaver cohorts (Hupka 2003; Glauser 2015). Finally, entering the labor
market without acquiring additional qualifications or not being in education, employ-
ment, or training (NEET) are rarely used alternatives.
Because the Swiss education system is highly stratified and tracking into different
school types after primary school already begins after 6th grade in most cantons, the
room for maneuver in making educational decisions towards the end of compulsory
education depends on the type of school attended. Educational opportunities for pupils
from the school type with basic requirements—i.e., the school type with the lowest cog-
nitive demands at lower secondary level—are mainly restricted to VET programs with
a shorter duration and/or lower cognitive requirements (EBA, EFZ), and these students
face the highest risk of accessing bridge year courses (Meyer 2003; Glauser 2015). In
contrast, pupils who attend a school type with advanced requirements formally have the
option to commence a wide range of VET programs (EBA, EFZ, FVB) or baccalaure-
ate and specialized schools, respectively. e vast majority of pupils from baccalaureate
preparation schools (pre-gymnasium) continue with general education programs (bac-
calaureate or specialized schools).
Compared to other OECD countries, a rather smooth school-to-work transition can
be observed for Switzerland (Blöchle etal. 2015; OECD 2015). e majority of youths
who have completed VET gain access to permanent, secure, and relatively well-paid
positions within a short timeframe (Müller and Schweri 2009; Glauser 2012; Lange etal.
2013). Due to the expansion of the universities of applied sciences and the long tradition
of the Swiss VET-system, a broad range of vocationally oriented further training is avail-
able at post-secondary and tertiary level. Nonetheless, the permeability between voca-
tionally and academically oriented tracks at secondary and tertiary level is low (Fazekas
and Field 2013, 71, SKBF 2014, 172). Since graduation from baccalaureate schools enti-
tles graduates to enroll at universities, those who opt for academically oriented tracks
and graduate from university on average earn more and have more opportunities to gain
access to higher social classes and to attain a more privileged way of life (Buchmann
etal. 2007; Müller etal. 2011; Schindler, 2014).
Page 4 of 25
Glauser and Becker Empirical Res Voc Ed Train (2016) 8:8
Theory andresearch
Educational pathways in an individual’s life course are embedded in varying societal
orders and social contexts, which, at the same time, systematically affect the duration
of participation in education and the opportunities to select and complete certain edu-
cational pathways (Coleman 1986; Becker and Hadjar 2013; Becker and Schulze 2013).
erefore, in order to explain the educational decisions made at the transition to upper
secondary education, individual, institutional, and regional constraints and opportuni-
ties need to be considered (see Fig.1). In this section, we first explicate how the nexus
of education and the employment system is generating institutionalized educational
pathways that are accompanied by specific opportunities and constraints with regard to
further education and working careers. Second, we elaborate on how disparities in the
social structure of educational attainment can be explained within a structural-individ-
ualistic theoretical framework. Finally, we clarify how educational aspirations and deci-
sions may be related to regional opportunities and constraints.
Impact ofthe institutional setting
In the Swiss context, educational pathways and trajectories at upper secondary and
subsequent levels are highly institutionalized, particularly those in which occupation-
specific skills are of greater importance. As in other countries with a system of “quali-
ficational space” (Maurice etal. 1986), a relatively high proportion of firms is engaged
in training apprentices and their organization of labor is aligned with qualifications
acquired in VET programs (Wolter et al. 2006). Furthermore, training companies
have a determining influence on the curricula of specific VET programs about branch
associations (Gonon 2012, 239). Albeit with a delay, changes in the business cycle,
the structure of the labor market, as well as the occupational structure likewise affect
the qualifications that are acquired in existing or newly created apprenticeships. Due
to the high degree of vocational specificity and standardization, firms can rely on the
Economic development
Labour markets
Occupaonal structure
and change
Social structure of
general educaon and
vocaonal educaon and
training
Educaonal system and supply of educaon and training
Firms and demand for occupaonal skills
(opportunies and restricons)
Class posion of
parental home
Educaonal
aspiraons
Educaonal
decision
School
performance
Fig. 1 Heuristic micro–macro model to explain the social structure of upper secondary education
Page 5 of 25
Glauser and Becker Empirical Res Voc Ed Train (2016) 8:8
acquired occupation-specific qualifications when filling vacant posts (Müller and Shavit
1998; Kerckhoff 2001; Wolter etal. 2006). is, in turn, implies that VET-qualifications
determine a candidate’s initial placement in the labor market, as well as the subsequent
opportunities related to continued professional development and occupational mobil-
ity between firms and occupations (Blossfeld 1985; Gangl 2001; Schmelzer 2012; Müller
and Schweri 2015). On the other hand, if pupils opt for VET, their chance of acquiring
a secure, qualified, and well-paid job is much higher compared to low-skilled workers
(Müller and Schweri 2009; Meyer and Bertschy 2011; Glauser 2012; Lange etal. 2013).
ese aspects, as well as the long tradition of VET, partly explain why vocational quali-
fications are well-respected in Switzerland, and why approximately two thirds of the
school leaver cohorts decide in favor of vocational training. Pupils’ feasible set of voca-
tional alternatives is restricted not only in terms of the type of school attended at lower
secondary level but the recruitment strategy and selection criteria of employers when
filling vacant apprenticeships. In this respect, the access to dual apprenticeships is a spe-
cial case of the ‘matching-problem’ between employers and employees at the point of
labor market entry (Müller etal. 2002).
In contrast, admission criteria and/or entrance examinations of school-based
apprenticeships, FVB, or general education programs are regulated at the cantonal
level. Although admission depends on school performance, the provision of baccalau-
reate schools, in German-speaking cantons in particular, has not changed much over
time. e matura rate in German-speaking Switzerland has remained relatively sta-
ble at approximately 20% in the last 15years and is approximately 8% lower than in
French- and Italian-speaking areas (Becker and Zangger 2013; BFS 2015; Glauser 2015).
e proportion of VET graduates within a canton is not only negatively correlated with
the probability of enrollment at baccalaureate and specialized schools but also with the
probability of attaining a tertiary degree (Imdorf et al. 2014). Pupils who fail to meet
the selection criteria of baccalaureate or specialized schools must consider other educa-
tional options, namely VET programs or bridge year courses.
Individual resources andrestrictions
In countries with a high degree of stratification and vocational specificity, a close rela-
tionship is observed between social origin and educational attainment, while intergen-
erational educational mobility is comparatively low (Pfeffer 2008; Müller and Kogan
2010; Falter and Wendelspiess ChávezJuárez 2011; Jann and Combet 2012; Falcon 2012;
Becker and Zangger 2013; Samuel etal. 2014). Children from socially underprivileged
families are less likely to succeed in directly securing a certifying upper secondary train-
ing place (Sacchi etal. 2011; Müller and Wolter 2014), and they are under-represented at
baccalaureate schools and the FVB (Schumann 2011; Stalder etal. 2011).
Social disparity with regard to lower and upper secondary educational attainment can
be explained to some extent within a structural individualistic theoretical framework
(Boudon 1974; Becker 2003; Stocké 2010; Kroneberg and Kalter 2012). Social inequalities
in educational attainment result from the correlation between social background and
school performance (primary effect of stratification). As a consequence of early track-
ing, during compulsory education, pupils from socially disadvantaged families are over-
represented in school types with lower requirements and, accordingly, in vocationally
Page 6 of 25
Glauser and Becker Empirical Res Voc Ed Train (2016) 8:8
oriented tracks at subsequent educational levels. However, IEO occurs even if school
performance is controlled for. is is referred to as the secondary effect of stratification
and describes the rationale behind educational decision-making, which varies between
actors from different social strata. All things being equal—as measured, for example, by
grade point averages (GPAs)—pupils from socially disadvantaged families are diverted
from academically oriented tracks and instead opt for VET. e reason is that students
from lower social classes, if they can choose between educational alternatives, tend to
prefer less demanding, time-consuming, costly, and risky educational tracks that never-
theless secure their access to qualified occupational positions and protect them against
unemployment (Müller and Pollak 2007; Breen etal. 2014). By contrast, educational
aspirations to continue with general education are higher among pupils from socially
privileged families because they otherwise face downward social mobility, the financial
burdens are lower, and they are familiar with general education (Erikson and Jonsson
1996; Breen and Goldthorpe 1997). ese theoretical assumptions are supported by
empirical results for different educational transitions and within a variety of educational
systems (representative of many: Stocké 2007; Erikson and Rudolphi 2010; Combet
2013; Karlson 2013; Becker etal. 2013; Glauser 2015).
Regional opportunity structures
So far, regional opportunity structures have received little attention in research on edu-
cational aspirations and decisions at the transition to upper secondary education. If
the impact of regional conditions of the labor market, the educational setting, or other
spatial aspects is considered, the focus has been on the transition from elementary to
lower secondary education (Sixt 2013; Zangger 2015), the decision to commence ter-
tiary education (Rephann 2002; Denzler and Wolter 2009; Reimer 2013), labor market
entry (Salvisberg and Sacchi 2014; Buchs etal. 2015), or the decision by firms to train
apprentices (Muhlemann and Wolter 2007). To our knowledge, the study from Weßling
etal. (2015) is the only one that focuses on regional labor market effects on opportuni-
ties to commence VET after compulsory education. With regard to Germany, the results
indicate that opportunities to commence VET are negatively correlated with the local
unemployment rate but uncorrelated with the labor market conditions of more remote
areas. However, what the authors cannot address with the SOEP-data is the question of
whether educational aspirations prior to the actual transition are likewise affected by
regional labor market conditions.
Regarding the decision to commence VET or general education, we assume that pupils
(and their parents) take the regional opportunity structures into account. e oppor-
tunity structures of the region in which they reside deliver information on opportuni-
ties and constraints related to different educational alternatives. is information, we
assume, influences their educational aspirations and final decisions. Variation in the
opportunity structures across regions should thus (partly) explain the variation observed
in pupils’ educational aspirations as well as the upper secondary track they attend
(hypothesis 1). e question as to whether these opportunity structures have an impact
on the educational attainment net of individual resources and the institutional setting of
the education system needs to be answered empirically.
Page 7 of 25
Glauser and Becker Empirical Res Voc Ed Train (2016) 8:8
Based on the discussion above, we argue that regional opportunity structures in Swit-
zerland differ with respect to the diversity and extent of the supply of educational alter-
natives. On the other hand, due to the strongly developed VET system and the strong
ties between the education system and labor market, the feasible set of educational alter-
natives depends on the condition as well as the structure of the regional labor market.
erefore, regions that differ in terms of occupational structure, relative importance of
labor market sectors, and dynamics of the labor market provide varying opportunity
structures. In regards to upper secondary educational attainment, we expect the follow-
ing: e more extensive the regional opportunity structures are, the higher the probabil-
ity that pupils opt for general education programs (hypothesis 2). We argue that regions
that can be characterized by more extensive opportunity structures provide a broader
range of not only vocationally but also academically oriented educational alternatives as
well as bridge year courses. Additionally, in these regions, the proportion of occupations
and workplaces in the third sector has significantly increased in the last few decades.
is development has deteriorated to some extent the employment prospects of persons
with vocational qualifications (Oesch 2013; Murphy 2014). e reason is that, when fill-
ing vacant posts that can be assigned to the third sector, firms prefer graduates from
school based training rather than graduates from dual VET, and their willingness to train
apprentices has decreased (Sheldon 2008; Salvisberg and Sacchi 2014). In contrast, due
to the trajectory of historical development, the widest range of apprenticeships exists,
even today, within the commercial and industrial sectors (Gonon 2012). Accordingly, in
regions in which the second sector of the labor market remains of great importance, the
opportunity structures may divert pupils from attending a general education program.
Methods
Data
To analyze the transition from lower to upper secondary education, we use recent data
from the DAB panel study (Determinants of educational choices and vocational training
opportunities). e focus of the DAB panel study is on the decision-making process and
educational pathways of pupils that left compulsory education in summer 2013. e data
are based on a stratified random sample of classes of 8th graders of the 2011/12 school
year within German-speaking Switzerland (for details, see (Glauser 2015), 125–132).
e survey time points were in the middle of the 8th grade (Jan/Feb 2012) as well as at
the beginning (Sept/Oct 2012) and end (May/June 2013) of the final year of compulsory
education. Data on the actual transition were collected in a follow-up study approxi-
mately 15months after pupils had left lower secondary education in October/Novem-
ber 2014. While in the first three waves the students were surveyed within their classes
using online questionnaires, the follow-up was conducted as an individual survey, using
a combination of online questionnaires and computer assisted telephone interviews
(CATI). At the class level, 215 to 199 out of the 296 classes sampled (72.6–67.2%) and,
at the individual level, approximately 3700–3300 students (response rates within waves:
90–96%) participated in the first three waves. Due to missing contact information, 2550
out of 3302 pupils who participated in wave 3 were invited to participate and 2237 pupils
(87.7%) participated in the follow-up. For the analysis, the sample was restricted to 2192
Page 8 of 25
Glauser and Becker Empirical Res Voc Ed Train (2016) 8:8
students who had participated in the follow-up and for whom information on educa-
tional attainment at upper secondary level was available.
To avoid loss of cases and hence statistical power, the data was multiply imputed using
chained equations (Allison 2001; White etal. 2011). Full information was available for
the following variables: the educational situation directly after as well as 15months after
completion of compulsory education, the scaling variable of the regional opportunity
structures, gender and attended school type at lower secondary level. e following vari-
ables are imputed (# missing values): GPA in German (160 Obs.) and mathematics (195
Obs.), parental social class (EGP; 165 Obs.) and educational level (361 Obs.) as well as
pupils’ educational aspirations from mid 8th grade (138 Obs.). Since we use grades in
standardized form in the analysis, the “just another variable” (JAV) approach is used
to impute missing values on GPA. Additionally, information on the sampling structure
(strata, class) was used for data imputation. e convergence of the imputation model
and the specifications of the number of iterations for the burn-in period is assessed
using trace line plots. We use 30 iterations for the burn-in period. e multivariate and
descriptive analyses are based on 50 imputed data sets.
To account for regional opportunity structures, we use data from the Swiss Federal
Statistical Office (FSO) at the MS-regional level. e total of 106 MS-regions (spatial
mobility regions; MS=mobilité spatiale) are characterized by a certain spatial homo-
geneity and reflect the principle of small—partially cross-cantonal—labor market areas
with functional orientation towards centers (Schuler and Joye 2000; Schuler etal. 2005;
BFS 2011). As the analyses are restricted to German-speaking cantons, information on
56 MS-regions was relevant and could be used (see Fig.3 in the Appendix for an over-
view of the MS-regions included in the analysis). If data were available at the community
level, those data have been aggregated at the MS-regional level.
Operationalization
e training places attended immediately after as well as 15months after leaving com-
pulsory education are used as dependent variables. ree educational outcomes are
differentiated: (1) VET programs including EBA, EFZ, and FVB; (2) continued general
education including specialized and baccalaureate schools; (3) bridge year courses that
do not lead to a certified upper secondary qualification (10th school year, au-pair, pre-
apprenticeships, preparatory courses, etc.). Descriptive statistics of the dependent and
independent variables are provided in the Appendix (see Tables4 and 5).
e following independent variables are used in the analysis. Pupils’ social background
is operationalized with the EGP-Class-Scheme (Erikson et al. 1979) and with paren-
tal education categorized according to the ISCED97-classification-scheme. Informa-
tion on the EGP-class, as well as the ISCED-level is based on students’ self-reporting
in wave 3. If information for both parents is available, the higher EGP-class was used.
Although the EGP-class of the parents and their attained educational degree could be
biased when based on students’ reporting, there is evidence that the reliability of child
reports on parental occupation is high, whilst this applies to a lesser extent to parental
educational level (Jerrim and Micklewright 2014; Engzell and Jonsson 2015). is is one
reason why parents’ reports on their occupational status and educational level, which
were also collected in wave 1, are not used. Another reason is that nonresponse and
Page 9 of 25
Glauser and Becker Empirical Res Voc Ed Train (2016) 8:8
item-nonresponse are lower when child reports are used. To account for the fact that
child reports on parental education are less reliable, a rather crude categorization is used
in which pupils whose parents had attained a tertiary degree are compared with those
whose parents had attained a lower educational level.
As control variables for school performance, we use GPA in German and mathemat-
ics. GPA information was provided by class teachers. Grades in Switzerland usually
rank from 1 to 6, 6 being the highest possible grade and 4 the minimum requirement.
Although school grades are no objective measure and may be biased, GPAs were used
for two reasons. First, for reasons of compliance with the participating teachers in the
DAB panel study, the implementation of an objective, teacher independent perfor-
mance test was resigned. Second, grades are the main selection criteria for admittance
to specialized and baccalaureate schools. Grades are used in z-standardized form in the
analyses. To account for institutional effects on educational attainment, the school type
attended at lower secondary level is used as a control variable.
Additionally, we control for gender as well as for pupils’ realistic educational aspira-
tions as measured in the mid 8th grade. e realistic educational aspirations are used as
a crude measurement of the secondary effects of stratification. is variable nonethe-
less reflects pupils’ subjective perception of the costs and benefits of different educa-
tional outcomes. Pupils were first asked what kind of upper secondary education they
would like to begin after 9th grade if they could decide solely according to their wishes
(idealistic educational aspiration). ey were then asked: “And if you are being realistic,
what do you think you will actually do after the 9th grade?”. Pupils’ responses to this
question are used to differentiate whether their aim was to begin VET, to continue with
general education, or to begin a bridge year course for the time being. We use informa-
tion on realistic aspirations from the mid 8th grade for two reasons. On the one hand,
realistic educational aspirations are influenced by the given structural constraints, e.g.
the regional opportunity structures (Paulus and Blossfeld 2007). On the other hand, a
high proportion of pupils who opt for general education enter pre-gymnasium already
after the 8th grade and will then continue with a baccalaureate school. For these pupils,
information on educational aspirations must be used from before the transition to
pre-gymnasium.
In the analyses, we use one z-standardized scaling variable to control for regional
opportunity structures. In the scaling variable, various aspects of the regional labor mar-
ket structures and conditions are reflected in terms of the total and the proportion of
workplaces in the second and third sector, the proportion of new jobs that have been
created within new firms in recent years, and the regional unemployment rate. Addition-
ally, we use the proportion of apprentices among employees, as well as the proportion of
persons entitled to enroll at universities in order to capture pull factors that might influ-
ence pupils to follow specific educational pathways. As a measure of the competition
among youths for high-demand training places, we use the youth ratio in the year 2000
as well as the change in the youth ratio within the years 2000–2010.
e scaling variable is the result of a principal-component factor analysis (PCF, see
Harman 1976, 136ff.) in which we finally used ten variables that were selected on the
basis of theoretical considerations and statistical quality criteria. Concerning the latter
Page 10 of 25
Glauser and Becker Empirical Res Voc Ed Train (2016) 8:8
aspect, the overall Kaiser-Meyer-Olkin measure of sampling adequacy (KMO=0.813),
as well as the KMO values of each of the variables used, are at least middling (KMO>0.8)
and Bartlett’s test of sphericity is highly significant. Based on the results of the parallel
analysis (see Horn 1965; Dinno 2009), one latent factor was extracted for which we pre-
sent the statistical quality criteria below (see Table1).
With the exception of the variable «proportion of work places in the second sector»,
all factor loadings are over 0.733, and communalities (=1 − uniqueness) are higher
than 0.538. Factor loadings are negative for three variables: «youth ratio», «proportion
of apprentices among all employees and apprentices» as well as «proportion of work
places in the second sector», whereas the factor loadings of the remaining variables are
positive. e average item-rest correlation and Cronbach’s
α
were computed with stand-
ardized items. e reliability coefficient of the generated scaling variable—as our meas-
urement of the regional opportunity structure—is excellent (
α=0.939
).
Overall, the highest values on the scaling variable are observed for the most urban
regions with a comparatively low (high) proportion of workplaces in the second (third)
sector and places in which most new jobs within new firms have been created in recent
years. Additionally, these regions are characterized by a rather low proportion of appren-
tices among employees but a high proportion of persons that are entitled to enroll at
universities. Besides a wide range of opportunities within the VET-system, these regions
offer the broadest supply of general education (see Fig.4 for a graphic representation of
the distribution of the scaling variable).
Statistical procedure
Since we analyze the educational situation immediately after leaving compulsory school-
ing as well as 15months afterwards, the analyses are conducted in two steps using mul-
tinomial logistic regression models (Long and Freese 2014). We use average marginal
effects (AMEs) to compare nested models and to minimize bias related to unobserved
heterogeneity (Mood 2010; Best and Wolf 2012). With regard to our dependent vari-
ables, AMEs provide an estimate of the population-averaged marginal effect of the inde-
pendent variables on the probability of attending a specific educational track.
Results
Educational situation afterleaving compulsory education
Immediately after compulsory education, the distribution of the dependent variable is
as follows: 63% of the school leaver cohort from summer 2013 had commenced VET,
21% a baccalaureate or specialized school, and 16% were unable to directly commence
with a certifying upper secondary track (see Table5). As expected, the institutional set-
ting strongly determines and channels educational attainment at upper secondary level
(see model 1 in Table2). Compared to pupils of the school type with basic requirements,
pupils of the school type with advanced requirements or those of the pre-gymnasium
have a significantly lower probability of commencing vocational education or a bridge
year course, whereas their probability of continuing with general education is increased.
e results related to effects at the individual level are in line with the findings of previ-
ous studies. Compared to their male counterparts, females’ probability of commencing
Page 11 of 25
Glauser and Becker Empirical Res Voc Ed Train (2016) 8:8
VET is approximately twenty percentage points lower. On the other hand, the results
indicate that on average—when controlling for GPA and social background—the pro-
portion of young females that continue with general education is higher compared to
males, but that the former also have difficulties in directly securing a certifying upper
secondary training place upon completion of compulsory schooling (see Glauser 2015,
142f.). Better grades in German and mathematics increase the probability of continuing
with general education, and protect against remaining without a certifying upper sec-
ondary training place.
e effects of parents’ EGP-class and their educational level indicate that IEO at this
educational transition can be observed even if the school type attended at lower sec-
ondary level and GPA are controlled. Pupils from socially underprivileged families
(EGP-classes V/VI/VII) compared to those from the upper service class (EGP-class I)
have a higher probability of commencing VET but a lower probability of continuing
with general education. e same is evident in respect of parental educational level. e
Table 1 Results ofthe principal-component factor analysis
a Source: Statatlas Switzerland, FSO; map‑id: 16, aggregated community level data based onFederal Population Census data
b Source: Statatlas Switzerland, FSO; map‑id: 15971, aggregated community level data based onFederal Population Census
data and Population and Households Statistics (STATPOP)
c Source: Statatlas Switzerland, FSO; map‑id: 17738, MS‑regional level data based on Structuralsurvey and Unemployment
Statistics (SECO)
d Source: Structural survey ; requested at FSO
e Source: MS‑regional level data based on Statistik der Unternehmensstruktur (STATENT) andStatistik der beruichen
Grundbildung (SBG), requested at FSO
f Source: Statatlas Switzerland, FSO; map‑id: 18182, aggregated community level data based onStatistik der
Unternehmensstruktur (STATENT )
g Source: Statatlas Switzerland, FSO; map‑id: 18185, aggregated community level data based onStatistik der
Unternehmensstruktur (STATENT )
h, i Source: Statatlas Switzerland, FSO; map‑id: 18182, 18185, aggregated community level databased on Statistik der
Unternehmensstruktur (STATENT )
j Source: Statatlas Switzerland, FSO; map‑id: 17735, MS‑regional level data based on BusinessCensus (BC) and Statistik der
Unternehmensdemograe (UDEMO)
Indicators ofregional opportunity structures Factor
loadings Unique-
ness Average item-
rest correlation
α
without
item i
Youth ratio, 2000a−0.856 0.267 0.813 0.929
Change in youth ratio between 2000 and 2010b0.826 0.318 0.777 0.931
Overall unemployment rate, 2013c0.738 0.456 0.671 0.936
Prop. of persons entitled to enroll at university on
all upper secondary qualifications, 2013d0.762 0.419 0.707 0.935
Prop. of apprentices among all employees and
apprentices, 2012e
−0.839 0.295 0.793 0.930
Total workplaces in the second sector, 2013f0.733 0.462 0.667 0.936
Total workplaces in the third sector, 2013g0.869 0.245 0.834 0.928
Prop. of workplaces in the second sector, 2013h−0.673 0.547 0.605 0.939
Prop. of workplaces in the third sector, 2013i0.884 0.218 0.849 0.928
Prop. of new jobs created in new firms between
2001–2012 among all new jobs in new firmsj0.846 0.285 0.803 0.930
Eigenvalue 6.488
Proportion of variance accounted for by factor 0.649
Average interitem correlation 0.605
Number of items in the scale 10
Cronbach’s
α
(standardized items) 0.939
Page 12 of 25
Glauser and Becker Empirical Res Voc Ed Train (2016) 8:8
probability of pupils from parents with a tertiary degree continuing with general educa-
tion is increased by approximately ten percentage points compared to pupils from par-
ents without a tertiary degree.
e empirical results support the assumption that regional opportunity structures
significantly affect pupils’ educational attainment. A one standard deviation increase in
the scaling variable reduces the probability that pupils have commenced VET by four
percentage points, whereas the probability that pupils have continued with general edu-
cation or commenced a bridge year course is increased by approximately 2% points.
Important, in our view, is the fact that we observe a negative effect on the probability
of pupils commencing VET and a positive effect in terms of continuing with general
education, respectively. e results indicate that regional opportunity structures have
an impact on the educational attainment net of individual resources and institutional
restrictions. If the regional opportunity structures offer a broader range of feasible
educational alternatives in terms of both vocationally as well as academically oriented
upper secondary tracks, and the total as well as the relative proportion of workplaces
in the third sector are comparatively high, then the probability of pupils commencing
VET decreases. In addition, the regional opportunity structures are positively correlated
with pupils’ probability of commencing bridge year courses. What cannot be addressed
is whether the results are indicating a stronger mismatch between the demand for and
the supply of specific training places in regions with broader opportunity structures, or
whether pupils, who reside in regions with a broader supply of such educational pro-
grams, make use of bridge year courses to improve their educational chances. In the
latter case, pupils from regions with restricted opportunity structures may be pulled
stronger to commence VET.
When we additionally control for the realistic educational aspirations of pupils
approximately 18months prior to completion of compulsory education, the explanatory
power of the model is improved (see model 2). e strong effects indicate that the realis-
tic educational aspirations are highly correlated with educational attainment during the
transition to upper secondary education. If pupils aspired in mid 8th grade to continue
with general education, then their probability of finally doing so is increased by 37%
points, whereas their probability of commencing VET is decreased by approximately
40% points. When realistic aspirations are considered, the effects of the school type
attended at lower secondary level remain significant in most cases, but the magnitude
of the effects declines substantially. is also applies partially to school performance.
Better grades in mathematics do not increase the probability of continuing with general
education but of commencing VET. e opposite is true in terms of grades in German.
Social origin effects are reduced but remain statistically significant. However, the lower
probability of commencing vocational education of pupils from parents with a tertiary
degree is insignificant. is likewise applies to the gender effect in terms of continuing
with general education. Unchanged is the finding that females’ probability of commenc-
ing bridge year courses is higher compared to men. Either the aim of young females is to
improve their chances of attaining a general education program, or there is a mismatch
between the supply of VET programs and the demands of young females.
With regard to the effects of the regional opportunity structures, only part of the direct
effects on educational attainment can be explained. In our view, the results nonetheless
Page 13 of 25
Glauser and Becker Empirical Res Voc Ed Train (2016) 8:8
Table 2 Educational situation directly afterleaving compulsory education [multinomial logistic regression, average marginal eects (AMEs)] (Source: DAB, own
calculations; estimates based on 50 imputed data sets)
Remarks: Discrete change eects for binary independent variables, robust standard errors in parentheses
Signicance levels: *p < 0.05, **p< 0.01, *** p < 0.001
Reference categories: a Basic requirements; b EGP‑class I; c ISCED level: no tertiary degree; d Vocational education and training
Model 1 Model 2
VET
EFZ/FVB Baccal. and
special. schools Bridge year
courses VET
EFZ/ FVB Baccal. and
special. schools Bridge year
courses
Regional opportunity structure −0.0430*** (0.0098) 0.0248*** (0.0072) 0.0181* (0.0075) −0.0305*** (0.0091) 0.0146* (0.0063) 0.0159* (0.0075)
Gender (Ref.: male) −0.1889*** (0.0204) 0.0595*** (0.0153) 0.1284*** (0.0159) −0.1409*** (0.0189) 0.0171 (0.0132) 0.1238*** (0.0157)
GPA German (z-standard.) −0.0452*** (0.0121) 0.0712*** (0.0103) −0.0260** (0.0092) −0.0218 (0.0111) 0.0419*** (0.0089) −0.0201* (0.0091)
GPA mathematics (z-standard.) 0.0189 (0.0119) 0.0324*** (0.0097) −0.0513*** (0.0089) 0.0347** (0.0106) 0.0125 (0.0081) −0.0473*** (0.0088)
School type 8th gradea
Advanced requirements −0.0459* (0.0204) 0.1809*** (0.0126) −0.1349*** (0.0178) −0.0235 (0.0213) 0.1402*** (0.0157) −0.1167*** (0.0185)
Pre-gymnasium −0.4812*** (0.0244) 0.6496*** (0.0235) −0.1685*** (0.0115) −0.3481*** 0.5040*** (0.0407) −0.1559*** (0.0145)
Social backgroundb,c
(0.0401)
EGP-class II 0.0424 (0.0334) −0.0107 (0.0238) −0.0318 (0.0274) 0.0358 (0.0307) 0.0020 (0.0197) −0.0378 (0.0270)
EGP-classes IIIa/b, IVa/b/c 0.0584 (0.0308) −0.0191 (0.0224) −0.0394 (0.0253) 0.0381 (0.0288) 0.0042 (0.0196) −0.0423 (0.0253)
EGP-classes V, VI, VIIa/b 0.1176*** (0.0335) −0.0671** (0.0249) −0.0505 (0.0266) 0.0921** (0.0313) −0.0460* (0.0217) −0.0461 (0.0267)
Tertiary degree (ISCED 5/6) −0.0698* (0.0274) 0.1065*** (0.0205) −0.0367 (0.0216) 0.0002 (0.0254) 0.0421* (0.0176) −0.0423* (0.0210)
Realistic educational aspiration (mid 8th grade)d
Baccalaureate/special. schools −0.4019*** (0.0289) 0.3674*** (0.0242) 0.0345 (0.0239)
Other −0.1927*** (0.0281) 0.0478* (0.0223) 0.1449*** (0.0234)
Observations/pseudo-
R2
2192/0.230 2192/0.341
Page 14 of 25
Glauser and Becker Empirical Res Voc Ed Train (2016) 8:8
indicate that the regional opportunity structures are perceived, subjectively evaluated,
and incorporated in pupils’ realistic educational aspirations. is assumption is sup-
ported by analyses in which the educational aspiration is used as a dependent variable
(see Table6 in the Appendix). A one standard deviation increase in the scaling variable
reduces the probability that pupils will aspire to attend VET, while the opposite holds
true in terms of aspiring to attend general education programs. Furthermore, there is no
direct effect of regional opportunity structures on the probability of pupils aspiring to
commence bridge year courses. e insignificant effect is not surprising, since the aim
of the pupils and their parents is to secure a certifying upper secondary training place.
Educational situation 15months afterleaving compulsory education
In the final step of the analysis, the focus is on the educational situation 15 months
after leaving compulsory education. e distribution of the dependent variable is as fol-
lows: 74% of the pupils in our sample attended VET, 21% a baccalaureate or specialized
school, and only 4% were unable to secure a certifying upper secondary training place
(see Table5). e vast majority of pupils who attended bridge year courses after leaving
compulsory education commenced VET. Almost unchanged is the proportion of pupils
that attended general education programs. e distribution is consistent with the find-
ings of other studies and administrative data (Hupka 2003; SKBF 2014).
Overall, the direction of the effects outlined in the previous section remains unchanged
(see model 1 in Table3). As before, we find strong effects of the school type attended at
lower secondary level on educational attainment 15months after leaving compulsory
education. e results confirm that better grades—particularly in German—reduce the
probability of pupils attending VET but increase the probability of them attending a gen-
eral education program. However, only better grades in mathematics significantly reduce
the probability of remaining without a certifying training place. As before, the gender
effects related to the probability of attending VET, baccalaureate schools and bridge year
courses are significant. With respect to social origins, the results confirm that pupils
from socially underprivileged families (EGP-classes V/VI/VII) have a higher probability
of attending VET and a lower probability of attending general education programs com-
pared to pupils from families of the upper service class (EGP-class I). It is no different in
relation to the educational level of the parents. e probability of pupils from parents
with a tertiary degree attending general education programs is increased by approxi-
mately 10% points compared to pupils from poorly educated social strata. ese results
demonstrate that within the Swiss educational system, the educational and thus in part
the occupational inheritance is evident (see Becker 1997; Becker and Schuchart 2010;
Lörz 2012). Pupils from parents that have obtained at most an upper secondary degree
and therefore, in the majority, vocational qualifications, are likely to tread the same edu-
cational path, although this does not necessarily imply that the same professional qualifi-
cations are acquired. is equally applies to pupils whose parents have attained a tertiary
degree. Since baccalaureate and specialized schools usually lead to matriculation certifi-
cates, a high proportion of these pupils will likewise attain a tertiary degree.
Finally, and as has been outlined in regards to the educational situation immedi-
ately after leaving compulsory schooling, the results confirm that the regional oppor-
tunity structures affect the educational attainment net of institutional constraints and
Page 15 of 25
Glauser and Becker Empirical Res Voc Ed Train (2016) 8:8
individual resources. Regional opportunity structures have a significant negative effect
in respect to the probability of attending VET and, conversely, a significant positive
effect on the probability of pupils attending a general education program. However, if
we additionally control for pupils’ realistic educational aspirations in model 2, the direct
effects of the opportunity structures are narrowed but still significant. e results differ
from those presented for the educational situation after leaving compulsory schooling,
in which the the positive effect on the probability of attending bridge year courses is sig-
nificant. Since only a minority of a school-leaver cohort is not able to realize a certifying
upper secondary training place within 15months after leaving compulsory education, it
is not surprising that institutional (attended school type at lower secondary level) and
individual constraints (GPA) matter more than the regional opportunity structure.
In line with the theoretical arguments related to secondary effects of stratification, the
effect of parental social class is explained completely, once we account for pupils’ realis-
tic educational aspirations. In contrast, parental education has a stronger direct effect on
pupils’ educational attainment. As reported previously, the gender effect related to the
probability of attending general education is no longer significant, indicating that part of
this effect is reflected within pupils’ educational aspirations and that females, for various
reasons, are more strongly diverted from VET while this educational pathway seems to
offer better prospects for males.
Discussion andconclusion
e aim of this paper was to address whether regional opportunity structures have an
impact on educational attainment net of institutional and individual restrictions and
how opportunity structures influence pupils’ educational attainment. In answering these
questions, we have used data from the DAB panel study enriched with administrative
data from the Swiss Federal Statistical Office at the MS-regional level. With this data, we
have analyzed the educational situation of pupils that left compulsory education in the
summer of 2013 in German-speaking Switzerland. In regards to educational attainment
immediately after and 15months after leaving compulsory schooling, we have differen-
tiated between VET, general education programs, and bridge year courses. To account
empirically for variation in the regional opportunity structures, we have constructed one
scaling variable based on principal component factor analysis. In the scaling variable,
various aspects of the regional labor market structures and conditions, as well as the
diversity and extent of the supply of educational alternatives, were considered.
Our findings indicate that regional opportunity structures have a direct effect on the
upper secondary track attended. In addition to the effects of the school type attended at
lower secondary level and individual constraints, we found that the more extensive the
regional opportunity structures are, the higher the probability of pupils attending gen-
eral education programs. In contrast, restricted opportunity structures increase pupils’
probability of attending VET programs. Moreover, the regional opportunity structures
are positively correlated with pupils’ probability of commencing bridge year courses
directly after leaving compulsory schooling. In our view, the varying results in regards
to educational attainment immediately after leaving compulsory schooling are related
to the fact that, at this stage, bridge year courses are of greater importance. Pupils from
regions with restricted opportunity structures have fewer options to commence bridge
Page 16 of 25
Glauser and Becker Empirical Res Voc Ed Train (2016) 8:8
Table 3 Educational situation 15 months afterleaving compulsory education [multinomial logistic regression, average marginal eects (AMEs)] (Source: DAB,
own calculations; estimates based on 50 imputed data sets)
Remarks: Discrete change eects for binary independent variables, robust standard errors in parentheses
Signicance levels: *p < 0.05, **p< 0.01, *** p < 0.001
Reference categories: a Basic requirements; b EGP‑class I; c ISCED level: no tertiary degree; d Vocational education and training
Model 1 Model 2
VET
EFZ/ FVB Baccal. and
special. schools Bridge year
courses VET
EFZ/ FVB Baccal. and
special. schools Bridge year
courses
Regional opportunity structure −0.0330*** (0.0082) 0.0249*** (0.0073) 0.0081 (0.0042) −0.0221** (0.0074) 0.0145* (0.0064) 0.0075 (0.0042)
Gender (Ref.: male) −0.0907*** (0.0173) 0.0636*** (0.0155) 0.0271** (0.0089) −0.0476** (0.0156) 0.0220 (0.0135) 0.0256** (0.0088)
GPA German (z-standard.) −0.0622*** (0.0112) 0.0671*** (0.0106) −0.0048 (0.0053) −0.0358*** (0.0099) 0.0393*** (0.0092) −0.0035 (0.0052)
GPA mathematics (z-standard.) −0.0128 (0.0110) 0.0334*** (0.0100) −0.0206*** (0.0056) 0.0049 (0.0096) 0.0145 (0.0084) −0.0194*** (0.0055)
School type 8th gradea
Advanced requirements −0.1517*** (0.0157) 0.1895*** (0.0126) −0.0378*** (0.0102) −0.1201** (0.0176) 0.1536*** (0.0154) −0.0335** (0.0103)
Pre-gymnasium −0.6071*** (0.0226) 0.6462*** (0.0223) −0.0391*** (0.0070) −0.4816*** (0.0378) 0.5163*** (0.0381) −0.0347*** (0.0083)
Social backgroundb,c
EGP-classes II 0.0259 (0.0277) −0.0327 (0.0240) 0.0068 (0.0155) 0.0149 (0.0245) −0.0209 (0.0203) 0.0060 (0.0153)
EGP-classes IIIa/b, IVa/b/c 0.0295 (0.0260) −0.0376 (0.0227) 0.0080 (0.0139) 0.0084 (0.0236) −0.0155 (0.0201) 0.0071 (0.0139)
EGP-classes V, VI, VIIa/b 0.0611* (0.0287) −0.0573* (0.0259) −0.0037 (0.0141) 0.0388 (0.0261) −0.0368 (0.0229) −0.0020 (0.0142)
Tertiary degree (ISCED 5/6) −0.1017*** (0.0241) 0.1033*** (0.0205) −0.0016 (0.0144) −0.0368 (0.0218) 0.0407* (0.0173) −0.0039 (0.0143)
Realistic educational aspiration (mid 8th grade)d
Baccalaureate/special. schools −0.3711*** (0.0262) 0.3580*** (0.0239) 0.0131 (0.0141)
Other −0.1028*** (0.0251) 0.0636** (0.0229) 0.0392** (0.0127)
Observations/pseudo-
R2
2192/0.247 2192/0.369
Page 17 of 25
Glauser and Becker Empirical Res Voc Ed Train (2016) 8:8
year courses and may, therefore, be forced to commence VET. We have argued in the
theoretical section that, on the one hand, regions with broader opportunity structures
provide a wide range of—vocational and general—education alternatives. On the other
hand, this aspect is correlated with the regional labor market structure and prospects of
labor market entry with occupation-specific or general qualifications. We could demon-
strate that pupils who reside in regions with rather restricted opportunity structures are
diverted from general education programs.
In addition, the analysis supports our assumption that pupils perceive and subjectively
evaluate regional opportunity structures, which thus organize pupils’ sets of feasible
actions. In this context, we could demonstrate that the regional opportunity structures
have an effect on pupils’ realistic educational aspirations. Once we control for educational
aspirations in the multivariate analyses, part of the direct effects of the opportunity struc-
tures on educational attainment after compulsory schooling can be explained. However,
since the effects of the regional opportunity structures remain significant this implies that
the real regional supply of educational opportunities structures the transitions to VET.
While the aspirations are significant push factors, the regional supply of vocational and
general educational alternatives could be isolated as pull factors for school leavers.
To our knowledge, this is the first study related to the impact of regional opportu-
nity structures on educational aspirations and educational attainment at the transition
to upper secondary education. Although we could demonstrate that educational aspi-
rations and educational attainment are correlated with opportunity structures, there
are several limitations of the study. First, the scaling variable that we used as a proxy
for opportunity structures is far from perfect. For example, we used the proportion
of apprentices among all employees and apprentices as a proxy for the supply of VET
programs, because, to date, data on apprenticeship openings are not available. It is of
utmost importance that data on the regional supply of apprenticeships are collected and
provided to the scientific community. Second, with the data and sample used, the find-
ings cannot be generalized to French- and Italian-speaking areas of Switzerland. ird,
since we observe the transition into upper secondary education for a single school-leaver
cohort only, the question of whether educational aspirations raise and fall with changes
in the opportunity structures cannot be addressed. Fourth, we did not control for selec-
tion effects, i.e., parents’ strategic choice of residence related to the supply of specific
educational alternatives. However, there is evidence that the probability for residential
mobility is increased by events such as gaining employment, marriage, or early family
planning, while residential mobility is less likely for families (Eluru etal. 2009). Finally, it
would be promising for further research to additionally account for variation of neigh-
boring regions to better capture the impact of the spatial structure of opportunity struc-
tures on educational outcomes.
Authors’ contributions
All authors read and approved the final manuscript.
Author details
1 Institute of Educational Science, University of Bern, Fabrikstrasse 8, 3012 Bern, Switzerland. 2 Department of Sociology
of Education, Institute of Educational Science, University of Bern, Fabrikstrasse 8, 3012 Bern, Switzerland.
Acknowledgements
We would like to thank Christoph Zangger, University of Bern, the guest editor Stephan Schumann, University of Kon-
stanz and the anonymous reviewers for their valuable advice during the elaboration of this article. Thank go to Riccarda
Neff and Adrien Làzàr for their assistance in the collection of data on the regional opportunity structures.
Page 18 of 25
Glauser and Becker Empirical Res Voc Ed Train (2016) 8:8
Competing interests
The authors declare that they have no competing interests.
Funding
The DAB panel study is substantially financed by the State Secretariat for Education, Research and Innovation (SERI). The
interpretations and conclusions are those of the authors and do not necessarily represent the views of the SERI.
Appendix
See Figs.2, 3 and 4 and Tables4, 5 and 6.
Fig. 2 Swiss educational system. Source: Swiss Conference of Cantonal Ministers of Education (EDK), see:
http://www.edk.ch/dyn/16833.php
Page 19 of 25
Glauser and Becker Empirical Res Voc Ed Train (2016) 8:8
Fig. 3 List of MS-regions
Page 20 of 25
Glauser and Becker Empirical Res Voc Ed Train (2016) 8:8
Fig. 4 Distribution of the scaling variable representing regional opportunity structures
Table 4 Descriptive statistics of (unstandardised) items used to construct the scaling
variable representing regional opportunity structures (N= 56 MS-regions) (Source: see
Table1)
Mean SD Min./max.
Youth ratio, year 2000 42.8 5.7 23.9/57.5
Change in youth ratio between 2000 and 2010 −7.9 4.4 −16.7/1.9
Unemployment rate, year 2013 2.4 0.7 0.5/4.3
Prop. of persons entitled to enrol at university among all
upper secondary qualifications, 2013 19.0 5.7 9.6/38.7
Prop. of apprentices among all employees and apprentices,
2012 6.3 1.1 3.8/8.9
Number of workplaces in the second sector, 2013 1340.8 768.8 96/2850
Number of workplaces in the third sector, 2013 7316.1 6529.7 399/38,909
Prop. of workplaces in the second sector, 2013 15.8 3.1 6.4/22.9
Prop. of workplaces in the third sector, 2013 72.3 11.2 38.7/93.5
Prop. of new jobs created in new firms between 2001–2012
among all new jobs in new firms 1.4 1.4 .03/9.3
Page 21 of 25
Glauser and Becker Empirical Res Voc Ed Train (2016) 8:8
Table 5 Descriptive statistics ofTables2 and3 (N=2192) (Source: DAB, own calculations;
estimates based on 50 imputed data sets)
Mean/
Proportion SD Min./max./
95% CI
Dependent variables
Educ. situation after leaving comp. educ.
Vocational education and training 0.625 0.605; 0.646
Baccalaureate/specialized school 0.213 0.195; 0.230
Bridge year courses 0.162 0.147; 0.177
Educ. situation 15 months after comp. educ.
Vocational education and training 0.744 0.726; 0.762
Baccalaureate/specialized school 0.213 0.196; 0.230
Bridge year courses 0.043 0.034; 0.051
Independent variables
Regional opportunity structure 0.219 0.981 −1.649/3.623
Gender: female 0.518 0.497; 0.539
School performance
GPA German 0.093 0.959 −6.206/2.974
GPA Mathematics 0.072 0.969 −5.434/2.731
School type 8th grade
Basic requirements 0.300 0.281; 0.319
Advanced requirements 0.578 0.557; 0.599
Pre-gymnasium 0.122 0.108; 0.136
EGP-class parents
EGP-class I 0.172 0.155; 0.188
EGP-class II 0.224 0.206; 0.243
EGP-classes IIIa/b, IVa/b/c 0.359 0.338; 0.380
EGP-classes V, VI, VIIa/b 0.245 0.226; 0.264
ISCED level parents
No tertiary degree (ISCED 2/4) 0.801 0.783; 0.819
Tertiary degree (ISCED 5/6) 0.199 0.181; 0.217
Educational aspiration in mid 8th grade
Vocational education and training 0.587 0.566; 0.608
Baccalaureate/specialized school 0.237 0.219; 0.255
Other 0.176 0.159; 0.192
Page 22 of 25
Glauser and Becker Empirical Res Voc Ed Train (2016) 8:8
Received: 6 November 2015 Accepted: 20 June 2016
References
Allison PD (2001) Missing data. Sage University Press series on quantitative applications in the social scienes, vol 07–136.
Sage, Thousand Oaks
BFS [Bundesamt für Statistik] (2011) Die Raumgliederungen der Schweiz 2011 - MS-Excel Version (be-b-00.04-rgs-01.xls).
http://www.bfs.admin.ch/bfs/portal/de/index/infothek/nomenklaturen/blank/blank/gemtyp/01.html, [Bundesamt
für Statistik]. Accessed 19 Sep 2014
BFS (2014) Bildungsabschlüsse Sekundarstufe II nach Sprachregion und Geschlecht, 1999–2013 (Dateien:
je-d-15.03.02.01.01 und je-d-15.03.02.02.01) 2014-08-15. http://www.bfs.admin.ch/bfs/portal/de/index/
themen/15/04/00/blank/uebersicht.html. Accessed 15 Aug 2014
BFS (2015) Maturitätsquote nach Maturitätstyp und Wohnkanton, 2014 (File: T405102d_d104_v8.xls). http://www.bfs.
admin.ch/bfs/portal/de/index/themen/15/17/blank/01.indicator.405102.4085.html?. Accessed 27 Oct 2015
Becker R (1997) Generationen und sozialer Wandel – eine Einführung. In: (Hg.) Generationen und sozialer Wandel. Gen-
erationsdynamik, Generationenbeziehungen und Differenzierung von Generationen. Leske + Budrich, Opladen, pp
9–22
Becker R (2003) Educational expansion and persistent inequalities of education. Utilizing subjective expected utility
theory to explain increasing participation rates in upper secondary school in the Federal Republic of Germany. Eur
Soc Rev 19:1–24. http://esr.oxfordjournals.org/cgi/content/abstract/19/1/1
Becker R, Hadjar A (2013) Gesellschaftliche Kontexte, Bildungsverläufe und Bildungschancen, Springer Fachmedien,
Wiesbaden, pp 511–553. doi:10.1007/978-3-531-18985-7_18
Becker R, Schuchart C (2010) Verringerung sozialer Ungleichheiten von Bildungschancen durch Chancenausgleich?
Ergebnisse einer Simulation bildungspolitischer Maßnahmen, Vierte veränderte Auflage, VS Verlag für Sozialwissen-
schaften, Wiesbaden, pp 413–436
Becker R, Schulze A (2013) Bildungskontexte. Springer Fachmedien, Strukturelle Voraussetzungen und Ursachen
ungleicher Bildungschancen, Wiesbaden
Becker R, Zangger C (2013) Die Bildungsexpansion in der Schweiz und ihre Folgen. Eine empirische Analyse des Wandels
der Bildungsbeteiligung und Bildungsungleichheiten mit den Daten der Schweizer Volkszählungen 1970, 1980,
1990 und 2000. Kölner Zeitschrift für Soziologie und Sozialpsychologie 65:423–449
Table 6 Determinants ofrealistic educational aspirations [multinomial logistic regression,
average marginal eects (AMEs)] Source: DAB, own calculations; estimates based on 50
imputed data sets
Remarks: Discrete change eects for binary independent variables, robust standard errors in parentheses
Signicance levels: * p<0.05, ** p<0.01, *** p<0.001
Reference categories: aBasic requirements; bEGP‑class I; cISCED level: no tertiary degree
VET
EFZ/FVB Baccal. and
special. schools Other
Regional opportunity
structure
−0.0352*** (0.0105) 0.0244** (0.0085) 0.0107 (0.0083)
Gender (Ref.: male) −0.1051*** (0.0219) 0.0845*** (0.0172) 0.0206 (0.0172)
School performance
GPA German (z-standard.) −0.0287* (0.0129) 0.0582*** (0.0113) −0.0295** (0.0098)
GPA mathematics
(z-standard.)
−0.0158 (0.0124) 0.0415*** (0.0108) −0.0257** (0.0096)
School type 8th gradea
Advanced requirements −0.0063 (0.0236) 0.1669*** (0.0175) −0.1606*** (0.0199)
Pre-gymnasium −0.3406*** (0.0328) 0.5033**** (0.0328) −0.1628*** (0.0141)
EGP-class parentsb
EGP-class II −0.0127 (0.0365) −0.0235 (0.0281) 0.0362 (0.0290)
EGP-classes IIIa/b, IVa/b/c 0.0328 (0.0335) −0.0378 (0.0253) 0.0050 (0.0265)
EGP-classes V, VI, VIIa/b 0.0610 (0.0356) −0.0392 (0.0290) −0.0218 (0.0275)
ISCED level parentsc
Tertiary degree (ISCED 5/6) −0.1653*** (0.0319) 0.1315*** (0.0247) 0.0338 (0.0276)
Observations/pseudo-
R2
2192/0.141
Page 23 of 25
Glauser and Becker Empirical Res Voc Ed Train (2016) 8:8
Becker R, Japel F, Beck M (2013) Diskriminierung durch Lehrpersonen oder herkunftsbedingte Nachteile von Migranten
im Deutschschweizer Schulsystem? Schweizerische Zeitschrift für Soziologie 39:517–549
Best H, Wolf C (2012) Modellvergleich und Ergebnisinterpretation in Logit- und Probit-Regressionen. Kölner Zeitschrift für
Soziologie und Sozialpsychologie 64:377–395. doi:10.1007/s11577-012-0167-4
Blossfeld HP (1985) “Berufseintritt und Berufsverlauf: Eine Kohortenananlyse über die Bedeutung des ersten Berufes in
der Erwerbsbiographie,” Mitteilungen aus der Arbeitsmarkt- und Berufsforschung, vol 18, pp 177–197. https://webvpn.
unibe.ch/mittab/1985/,DanaInfo=doku.iab.de+1985_2_MittAB_Blossfeld.pdf
Blöchle SJ, Flake R, Khairi-Taraki T, Körbel M, Pierenkemper S, Rauland C, Werner D, Wörndl D (2015) Berufsausbildung
für Europas Jugend. Voneinander lernen, miteinander gestalten (Studie zusammen mit der Hans-Böckler-Stiftung,
der Konrad-Adenauer-Stiftung und der Vodafone Stiftung Deutschland), Technical report, Institut der deutschen
Wirtschaft Köln
Boudon R (1974) Education, opportunity and social inequality. Wiley, Changing Prospects in Western Society, New York
Breen R, Goldthorpe JH (1997) Explaining educational differentials. Towards a formal rational action theory. Ration Soc
9:275–305
Breen R, van de Werfhorst HG, Jæger MM (2014) Deciding under doubt. A theory of risk aversion, time discounting pref-
erences, and educational decision-making. Eur Soc Rev 30:258–270. http://esr.oxfordjournals.org/content/30/2/258.
abstract
Buchmann M, Sacchi S, Lamprecht M, Stamm H (2007) Tertiary education expansion and social inequality in Switzerland.
Stanford University Press, Stanford, pp 321–348
Buchs H, Müller B, Buchmann M (2015) Qualifikationsnachfrage und Arbeitsmarkteintritt in der Schweiz. Kölner Zeitschrift
für Soziologie und Sozialpsychologie 67:709–736. doi:10.1007/s11577-015-0342-5
Coleman JS (1986) Social-theory, social-research, and a theory of action. Am J Soc 91:1309–1335. http://www.jstor.org/
stable/2779798
Combet B (2013) Zum Einfluss von primären und sekundären Effekten der sozialen Herkunft beim zweiten schulischen
Übergang in der Schweiz. Ein Vergleich unterschiedlicher Dekompositions- und Operationalisierungsmethoden. Ein
Vergleich unterschiedlicher Dekompositions- und Operationalisierungsmethoden. Schweizerische Zeitschrift für
Bildungswissenschaften 35:447–471
de Lange M, Gesthuizen M, Wolbers MHJ (2013) Youth labour market integration across Europe. Eur Soc 16:194–212. doi:
10.1080/14616696.2013.821621
Denzler S, Wolter SC (2008) Selbstselektion bei der Wahl eines Lehramtsstudiums. Zum Zusammenspiel individueller und
institutioneller Faktoren. Beiträge zur Hochschulforschung 30:112–141
Dinno A (2009) Implementing Horn’s parallel analysis for principal component analysis and factor analysis. Stata J
9:291–298. http://www.stata-journal.com/article.html?article=st0166
Eluru N, Sener I, Bhat C, Pendyala R, Axhausen K (2009) Understanding residential mobility. Transp Res Rec J Transp Res
Board 2133:64–74. http://trrjournalonline.trb.org/doi/abs/10.3141/2133-07
Engzell P, Jonsson J (2015) Estimating social and ethnic inequality in school surveys. Biases from child misreporting and
parent nonresponse. Eur Soc Rev 31:312–325. http://esr.oxfordjournals.org/content/31/3/312.abstract
Erikson RJJ (1996) Explaining class inequality in education: the Swedish test case. Westview Press, Boulder, pp 1–63
Erikson R, Rudolphi F (2010) Change in social selection to upper secondary school. Primary and secondary effects in
Sweden. Eur Soc Rev 26:291–305. http://esr.oxfordjournals.org/content/26/3/291.abstract
Erikson R, Goldthorpe JH, Portocarero L (1979) Intergenerational class mobility in three western European societies.
England, France and Sweden. Br J Soc 30:415–441. http://www.jstor.org/stable/589632
Falcon J (2012) Temporal trends in intergenerational social mobility in Switzerland. A cohort study of men and women
born between 1912 and 1974. Swiss J Soc 38:153–175
Falter JM, Chávez Juárez F (2011) Does tracking shape intergenerational transmission of educational attainment? Evi-
dence from Switzerland (Working-paper). http://ssrn.com/abstract=1938281. Accessed 19 Sept 2014
Fazekas M, Field S (2013) A skills beyond school review of Switzerland (Skills OECD: building the right skills and turning
them into better jobs and better lives). OECD, Paris
Gangl M (2001) European patterns of labour market entry. A dichotomy of occupationalized vs. non-occupationalized
systems? Eur Soc 3:471–494. doi:10.1080/14616690120112226
Glauser D (2012) Erwerbseintritt von gering qualifizierten Jugendlichen in der Schweiz: Unqualifizierte Beschäftigung
und Einkommensnachteile im Vergleich zu Jugendlichen mit Berufsbildungsabschluss. Beltz Juventa, Weinheim, pp
231–250
Glauser D (2015) Berufsausbildung oder Allgemeinbildung. Soziale Ungleichheiten beim Übergang in die Sekundarstufe
II in der Schweiz. Springer VS, Wiesbaden. doi:10.1007/978-3-658-09096-8_2
Gonon P (2012) Entstehung und Dominanz der dualen Berufsausbildung in der Schweiz. Springer Fachmedien, Wies-
baden, pp 221–242. doi:10.1007/978-3-531-19071-6_11
Harman HH (1976) Modern factor analysis. University of Chicago Press, Chicago
Horn JL (1965) A rationale and test for the number of factors in factor analysis. Psychometrika 30:179–185. doi:10.1007/
BF02289447
Hupka S (2003) Ausbildungssituation und -verläufe. Übersicht. BFS, Neuchâtel
Imdorf C, Sacchi S, Wohlgemuth K, Cortesi S, Schoch A (2014) How cantonal education systems in Switzerland promote
gender-typical school-to-work transitions. Schweizerische Zeitschrift für Soziologie 40:175–196
Jann B, Combet B (2012) Zur Entwicklung der intergenerationalen Mobilität in der Schweiz. Swiss J Soc 38:177–199
Jerrim J, Micklewright J (2014) Socio-economic gradients in children’s cognitive skills. Are cross-country comparisons
robust to who reports family background? Eur Soc Rev 30. http://esr.oxfordjournals.org/content/early/2014/08/25/
esr.jcu072.abstract
Karlson KB (2013) Summarizing primary and secondary effects. Res Soc Stratif Mobil 33:72–82. http://www.sciencedirect.
com/science/article/
Kerckhoff AC (2001) Education and social stratification processes in comparative perspective. Soc Educ 74:3–18. http://
www.jstor.org/stable/2673250
Page 24 of 25
Glauser and Becker Empirical Res Voc Ed Train (2016) 8:8
Kroneberg C, Kalter F (2012) Rational choice theory and empirical research. Methodological and theoretical contributions
in Europe. Annu Rev Soc 38:73–92. doi:10.1146/annurev-soc-071811-145441
Long JS, Freese J (2014) Regression models for categorical dependent variables using Stata, 3rd edn. Stata Press, College
Station
Lörz M (2012) Mechanismen sozialer Ungleichheit beim Übergang ins Studium: Prozesse der Status- und Kulturre-
produktion. vol 52 of Kölner Zeitschrift für Soziologie und Sozialpsychologie Sonderhefte, Springer Fachmedien,
Wiesbaden, pp 302–324. doi: 10.1007/978-3-658-00120-9_13
Maurice M, Sellier F, Silvestre JJ (1986) The social foundations of industrial power. MIT Press, A comparison of France and
Germany, Cambridge
Meyer T (2003) Zwischenlösung - Notlösung?. BFS, Neuchâtel
Meyer T, Bertschy K (2011) The long and winding road from education to labour market. The TREE cohort six years after
leaving compulsory school. Seismo, Zürich, pp 92–119
Mood C (2010) Logistic regression. Why we cannot do what we think we can do, and what we can do about it. Eur Soc
Rev 26:67–82. http://esr.oxfordjournals.org/content/26/1/67.abstract
Murphy EC (2014) Workers’ movement out of declining occupations in Great Britain, Germany and Switzerland. Eur Soc
Rev 30:685–701. http://esr.oxfordjournals.org/content/early/2014/08/09/esr.jcu066.abstract
Mühlemann S, Wolter SC (2007) Regional effects on employer provided training: evidence from apprenticeship training
in Switzerland. Zeitschrift für ArbeitsmarktForschung 40:135–147
Müller W, Kogan I (2010) Education. Springer, New York, pp 217–289. doi:10.1007/978-0-387-88199-7_9
Müller W, Pollak R (2007) Weshalb gibt es so wenige Arbeiterkinder in Deutschlands Universitäten?. VS Verlag für Sozial-
wissenschaften, Wiesbaden, pp 303–342
Müller B, Schweri J (2009) Berufswechsel beim Übergang von der Lehre in den Arbeitsmarkt. Schweizerische Zeitschrift
für Bildungswissenschaften 31:199–227
Müller B, Schweri J (2015) How specific is apprenticeship training? Evidence from inter-firm and occupational mobility
after graduation. Oxf Econ Papers. http://oep.oxfordjournals.org/content/early/2015/06/27/oep.gpv040.abstract
Müller W, Shavi Y (1998) The institutional embeddedness of the stratification process. A comparative study of qualifica-
tions and occupations in thirteen countries. Clarendon Press, Oxford
Müller B, Wolter SC (2014) The role of hard-to-obtain information on ability for the school-to-work transition. Empir Econ
46:1447–1474. doi:10.1007/s00181-013-0709-2
Müller W, Gangl M, Scherer S (2002) Übergangsstrukturen zwischen Bildung und Beschäftigung. Juventa, Weinheim
Müller W, Pollak R, Reimer D, Schindler S (2011) Hochschulbildung und soziale Ungleichheit. VS Verlag für Sozialwissen-
schaften, Wiesbaden, pp 289–327. doi:10.1007/978-3-531-92759-6_11
OECD (2015) OECD Publishing, Paris, doi: 10.1787/eag-2015-83-en
Oesch D (2013) Occupational change in Europe. Oxford University Press, How technology and education transform the
job structure, Oxford
Paulus W, Blossfeld HP (2007) Schichtspezifische Präferenzen oder sozioökonomisches Entscheidungskalkül? Zur Rolle
elterlicher Bildungsaspirationen im Entscheidungsprozess beim übergang von der Grundschule in die Sekunda-
rstufe. Zeitschrift für Pädagogik 53:491–508
Pfeffer FT (2008) Persistent inequality in educational attainment and its institutional context. Eur Soc Rev 24:543–565.
http://esr.oxfordjournals.org/content/24/5/543.abstract
Reimer D (2013) Kontexteffekte und soziale Ungleichheit beim übergang von der Schule zur Hochschule. Springer Fach-
medien Wiesbaden, Wiesbaden, pp 405–429. doi:10.1007/978-3-531-18985-7_14
Rephann TJ (2002) The importance of geographical attributes in the decision to attend college. Socio Econ Plan Sci
36:291–307. http://www.sciencedirect.com/science/article/
Sacchi S, Hupka-Brunner S, Stalder BE, Gangl M (2011) Die Bedeutung von sozialer Herkunft und Migrationshintergrund
für den Übertritt in anerkannte nachobligatorische Ausbildungen in der Schweiz. Seismo, Zürich
Salvisberg A, Sacchi S (2014) Labour market prospects of Swiss career entrants after completion of vocational education
and training. Eur Soc 16:255–274. doi:10.1080/14616696.2013.821623
Samuel R, Bergman MM, Hupka-Brunner S (2014) Longitudinal effects of social background on educational and occupa-
tional pathways within early and strong school tracking. Longitud Life Course Stud 5:1–18
Schindler S (2014) Wege zur Studienberechtigung-Wege ins Studium?. Springer Fachmedien, Wiesbaden.
doi:10.1007/978-3-658-03841-0
Schmelzer P (2012) The consequences of job mobility for future earnings in early working life in Germany. Placing indi-
rect and direct job mobility into institutional context. Eur Soc Rev 28:82–95
Schuler M, Dessemontet P, Joye D (2005) Die Raumgliederung der Schweiz. BFS, Neuchâtel
Schuler M, Dominique JO (2005) Typologie der Gemeinden der Schweiz: 1980–2000. BFS, Neuchâtel
Schumann S (2011) Leistungs-und Herkunftseffekte beim Hochschulzugang in der Schweiz. Ein Vergleich zwischen
Absolventinnen und Absolventen mit gymnasialer Maturität und mit Berufsmaturität. Zeitschrift für Pädagogik
57:246–268
Sheldon G (2008) Die Rolle der Berufsbildung in der Bekämpfung des Fachkräftemangels, Technical report. Universität
Basel, Forschungsstelle für Arbeitsmarkt und Industrieökonomik
Sixt M (2013) Wohnort, Region und Bildungserfolg. Die strukturelle Dimension bei der Erklärung von regionaler Bildung-
sungleichheit. Springer Fachmedien Wiesbaden, Wiesbaden, pp 457–481. doi:10.1007/978-3-531-18985-7_16
SKBF [Schweizerische Koordinationsstelle fur Bildungsforschung] (2014) Bildungsbericht Schweiz 2014. SKBF, Aarau
Stalder BE, Meyer T, Hupka-Brunner S (2011) Leistungsschwach–bildungsarm? Ergebnisse der TREE-Studie zu den PISA-
Komepetenzen als Prädiktoren für Bildungschancen in der Sekundarstufe II. Seismo, Zärich, pp 201–216
Stocké V (2007) Explaining educational decision and effects of families social class position. An empirical test of the Breen
Goldthorpe model of educational attainment. Eur Soc Rev 23:505–519. http://esr.oxfordjournals.org/cgi/content/
abstract/23/4/505
Stocké V (2010) Der Beitrag der Theorie rationaler Entscheidung zur Erklärung von Bildungsungleichheit. VS Verlag für
Sozialwissenschaften, Wiesbaden. doi:10.1007/978-3-531-92576-9_4
Page 25 of 25
Glauser and Becker Empirical Res Voc Ed Train (2016) 8:8
Weßling K, Hartung A, Hillmert S (2015) Spatial structure counts. The relevance of regional labour-market conditions for
educational transitions to vocational training. Empir Res Vocat Educ Train 7:1–20. doi:10.1186/s40461-015-0024-6
White IR, Royston P, Wood AM (2011) Multiple imputation using chained equations: issues and guidance for practice. Stat
Med 30:377–399
Wolter SC, Mühlemann S, Schweri J (2006) Why some firms train apprentices and many others do not. Ger Econ Rev
7:249–264. doi:10.1111/j.1468-0475.2006.00155.x
Zangger C (2015) The social geography of education. Neighborhood, class composition, and the educational achieve-
ment of elementary school students in Zurich, Switzerland. Zeitschrift für Soziologie 44:292–310
