Content uploaded by Markus Klein
Author content
All content in this area was uploaded by Markus Klein on Feb 24, 2015
Content may be subject to copyright.
Edinburgh Research Explorer
The Impact of Social Origin on Graduates' Early Occupational
Destinations
Citation for published version:
Jacob, M, Klein, M & Iannelli, C 2015, 'The Impact of Social Origin on Graduates' Early Occupational
Destinations: An Anglo-German Comparison' European Sociological Review., 10.1093/esr/jcv006
Digital Object Identifier (DOI):
10.1093/esr/jcv006
Link:
Link to publication record in Edinburgh Research Explorer
Document Version:
Publisher final version (usually the publisher pdf)
Published In:
European Sociological Review
Publisher Rights Statement:
© The Author 2015. Published by Oxford University Press.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License
(http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in
any medium, provided the original work is properly cited.
General rights
Copyright for the publications made accessible via the Edinburgh Research Explorer is retained by the author(s)
and / or other copyright owners and it is a condition of accessing these publications that users recognise and
abide by the legal requirements associated with these rights.
Take down policy
The University of Edinburgh has made every reasonable effort to ensure that Edinburgh Research Explorer
content complies with UK legislation. If you believe that the public display of this file breaches copyright please
contact openaccess@ed.ac.uk providing details, and we will remove access to the work immediately and
investigate your claim.
Download date: 24. Feb. 2015
The Impact of Social Origin on Graduates’ Early
Occupational Destinations—An Anglo-German
Comparison
Marita Jacob,
1,*
Markus Klein
2
and Cristina Iannelli
2
1
University of Cologne, Institute of Sociology and Social Psychology (ISS), Greinstr. 2, 509309 Ko¨ln, Germany
and
2
University of Edinburgh, Moray House School of Education, St. John’s Land, EH8 8AQ Edinburgh, UK
*Corresponding author. Email: marita.jacob@uni-koeln.de
Submitted April 2014; revised October 2014; accepted December 2014
Abstract
This article examines the impact of social origin on tertiary graduates’ labour market outcomes
in Germany and the United Kingdom, two distinct countries in terms of higher education systems,
labour market structures, and their linkages. Data from the 2005 REFLEX survey, OLS regression and
linear probability models are used to analyse the effect of parental education on graduates’
occupational destinations at two time points: at labour market entry and five years after graduation.
We test various hypotheses on country variation (i) in the strength of association between origin and
occupational destinations, (ii) in the mechanisms by which social origin affects occupational destin-
ations (i.e. via qualitative education differences), and (iii) in the extent to which social origin matters at
different career stages. The results show that parental education effects are similar in the two coun-
tries when occupational destinations are analysed using the International Socio-Economic Index of
Occupational Status (ISEI). They substantially differ when the analyses focus on entry into the higher-
service class. In this latter case, both the gross and the net effects of parental education are stronger
in the United Kingdom than in Germany. However, country differences in parental education effects
reduce when graduates’ occupational outcomes are analysed 5 years after graduation.
Introduction
Several studies in social stratification research have
shown that social origin influences a child’s occupa-
tional position, and that this effect holds over and above
educational attainment (Breen, 2004;Iannelli and
Paterson, 2007). However, in some countries the
influence of social class origin on social class destin-
ations is weaker among the highly educated than
among the lower educated (Hout, 1988;Iannelli and
Paterson, 2007;Breen and Jonsson, 2008). This com-
parison between educational groups indicates that the
acquisition of a tertiary qualification may reduce social
class differences in occupational outcomes and increases
the chances for social mobility. Social stratification re-
searchers warn, however, that differentiation within the
education system may continue to reproduce within-
group differences: even when students from lower
classes achieve access to tertiary education, qualitative
differences, such as field of study and type of institution
attended, matter in the reproduction of social inequal-
ities (Lucas, 2001, called this process ‘Effectively
Maintained Inequality’). In our article, we look at
V
CThe Author 2015. Published by Oxford University Press.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/),
which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.
European Sociological Review, 2015, 1–13
doi: 10.1093/esr/jcv006
Original Article
European Sociological Review Advance Access published February 9, 2015
at Edinburgh University on February 18, 2015http://esr.oxfordjournals.org/Downloaded from
within-group differences among tertiary graduates in
Germany and the United Kingdom. We focus on the
most educated individuals (i) to examine whether and to
what extent social origin still matters for labour market
outcomes in this group, and (ii) to analyse the mechan-
isms by which social origin may influence labour market
allocation net of educational attainment.
Moreover, by comparing Germany and the United
Kingdom, we gain a better understanding of the role of
national institutional differences in shaping the transi-
tion from education to the labour market (e.g. Breen,
2010;Triventi, 2013). Social origin might be particu-
larly relevant if competition among graduates is fierce
and if there is an absence of tight links between the
higher education (HE) qualification achieved and the
labour market destination. In such a situation, the fam-
ily of origin may provide useful resources which help
graduates from more privileged backgrounds to acquire
higher status and remunerative occupations. We test this
idea by comparing Germany and the United Kingdom,
two countries that show considerable differences in the
degree of signalling capacity of education, competition
among graduates, and occupational-specific allocation.
We further pursue a life course perspective, looking
at both labour market entry and early career develop-
ment five years later. Distinguishing between the
position at immediate labour market entry and develop-
ments in early working careers has been proven to be
fruitful, particularly in comparative research (Scherer,
2001;Jacob and Weiss, 2011). Assuming that one’s own
labour market-related resources and work experience
gain importance in people’s occupational career, social
origin effects visible at labour market entry may de-
crease across the career and country differences in social
differentials become smaller.
Against this background, the article investigates the
following questions:
1. Does social origin affect graduates’ outcomes at
labour market entry in Germany and the United
Kingdom? If so, do Germany and the United
Kingdom vary in the extent to which social origin
matters?
2. Do qualitative differences, such as the field of study
chosen and the type of HE institution attended,
explain the social origin effect more in one country
than in the other?
3. Do country differences in the effect of social origin
persist five years after graduation?
We discuss potential mechanisms that may explain
the impact of social origin on labour market outcomes
among tertiary graduates in our theoretical
considerations. The following section describes differ-
ences in the institutional setting of HE and graduate la-
bour market structures between Germany and the
United Kingdom. Thereafter, we formulate hypotheses
on country differences in social origin effects on gradu-
ates’ occupational outcomes. Subsequently, we present
data, methods, and the empirical results. In the conclud-
ing section, we will discuss the results in light of the for-
mulated hypotheses.
Class of Origin and Graduates’ Labour
Market Outcomes: Theoretical
Considerations
Labour market allocation can be thought of as a process
of matching applicants to vacancies that depends upon
applicants’ assets, the degree of job competition, and
employers’ hiring decisions. Various theories have been
proposed to explain this process. Particularly relevant
for this article are two sets of theories. Credentialist and
signalling theories focus on meritocratic mechanisms
of job allocation that would predict little or no effect of
social origin among HE graduates. Social reproduction
theories, on the other hand, would predict the persist-
ence of social inequalities in graduates’ labour market
outcomes due to institutional differentiation within the
HE sector and due to the role that parental resources
continue to exercise among graduates.
Credentialist and Signalling Theories
According to credentialist approaches, allocation into
the labour market is organized by credentials (e.g.
Collins, 1979). From this perspective, formal educa-
tional qualifications and occupational positions are
closely linked and serve occupational and social closure.
Credentialist allocation may be particularly salient in
the graduate labour market. On the one hand, the trad-
itional professions collectively maintain barriers by legal
constraints of certification and fine traditions (Bol and
Van de Werfhorst, 2011). On the other hand, allocation
into the public sector and into large-scale organizations
is highly formalized, managed, and bureaucratically
organized by personnel departments that consider only
educational levels for choosing among applicants
(Mastekaasa, 2011). Hence, individual ascriptive char-
acteristics—among them social background—should
have no or only a minor effect on labour market out-
comes. National labour markets can be more or less
organized by credentials: the stronger the linkages be-
tween educational qualifications and occupational
2European Sociological Review, 2015, Vol. 0, No. 0
at Edinburgh University on February 18, 2015http://esr.oxfordjournals.org/Downloaded from
positions, the less (social) variation in graduates’ labour
market outcomes exists.
Signalling approaches assume that education serves as
a signal of expected future productivity (e.g. Spence,
1973;Stiglitz, 1975). The signalling capacity of education
varies between education systems according to ‘stratifica-
tion’, i.e. the extent to which secondary education is div-
ided into distinct tracks, and ‘standardization’, i.e. the
degree of uniformity in the school-leaving examination,
curricula, HE entry requirements, and other educational
features. In highly stratified and standardized education
systems, qualifications are more reliable signals for em-
ployers than in less stratified and standardized systems.
Employers might further assume that unobserved charac-
teristics such as motivation, the ability to learn, or occu-
pational aspirations are at a similar level among
graduates who have already mastered several previous
transitions and exams. This equalizing effect of gradu-
ation is particularly strong if previous transitions have
been (socially) selective (Mare, 1980). The credentalist
and signalling arguments particularly matter at the begin-
ning of the working career when unknown applicants are
hired for the first time whereas, later in one’s career, ob-
servable productivity and working experience become
more important for promotion or job mobility.
Social Reproduction Theories
Social reproduction theorists argue that, despite educa-
tional expansion and the significant role education has
for job allocation, social origin continues to matter. This
is because individuals from more advantaged family back-
grounds are able to adopt new strategies to ensure that
their offspring will maintain their status. Several studies
have shown evidence for such qualitative differences
within the education system. For instance, students from
different classes vary in their educational pathways and
attainment prior to HE (Iannelli, Gamoran, and Paterson,
2011;Schindler, 2014), the chosen field of study (Van de
Werfhorst, Sullivan, and Cheung, 2003;Reimer and
Pollak, 2010), the chosen type of tertiary institution
(Boliver, 2011;Iannelli, Gamoran, and Paterson, 2011;
Schindler and Reimer, 2011), or labour market participa-
tion during their studies (Jacob and Weiss, 2012). These
differences may lead to diverse outcomes among stu-
dents with the same level of education but different origin
class, so-called Effectively Maintained Inequality in edu-
cation (Lucas, 2001). Hence, part of social inequalities in
graduates’ labour market outcomes might be due to these
qualitative differences. From this perspective, a more in-
clusive system of HE may reinforce the importance of
qualitative distinctions within HE and reproduce social
inequalities.
Besides formal qualifications, students may take an
advantage from parental resources, such as cultural and
social capital that are directly or indirectly transmitted
from parents to children (Bourdieu, 1986;Coleman,
1988). However, the extent to which cultural and social
resources are important may depend on employers’ de-
mands and the labour market structure. If competition
among graduates for favourable labour market positions
is high, any other differentiating factor beyond the de-
gree may serve as a competitive advantage. Endowment
with social and cultural capital and above all soft skills
might be particularly helpful for entering high manager-
ial positions or services (Jackson, Goldthorpe, and
Mills, 2005). This direct influence of parental back-
ground via cultural and social capital is assumed to be
particularly relevant for early labour market allocation,
but to a lesser extent later in the life course as own
resources accumulate during one’s working life.
An Anglo-German Comparison
In the following, we discuss both sides of the institu-
tional conditions of the matching process (educational
system and labour market structures) and their linkages
to deriving hypotheses on country differences in the allo-
cation process of graduates into the labour market and
early career development.
Educational System: Selectivity and
Differentiation
In the highly differentiated tripartite system of
Germany, pupils are selected into different school tracks
early, commonly at the age of 10 in most La¨nder. Access
to HE is determined by the acquisition of the Abitur, the
upper-secondary school graduation certificate. Among
those who are entitled to enter HE, a considerable num-
ber of students prefer to enrol in the dual system of
apprenticeship (Schindler and Reimer, 2011). Due to the
early selection points and the ‘diversion’ into vocational
training, selectivity of tertiary students in Germany is
comparatively high.
In the United Kingdom, selectivity of graduates is
much weaker than in Germany: secondary education in
the United Kingdom mostly comprises comprehensive
schools where pupils are taught together until the age of
16.
1
The low selectivity in the United Kingdom is further
strengthened by the fact that there are no formal restric-
tions to accessing upper-secondary education at the end
of compulsory schooling.
With regard to educational expansion, a much higher
percentage of young people graduate from tertiary
European Sociological Review, 2015, Vol. 0, No. 0 3
at Edinburgh University on February 18, 2015http://esr.oxfordjournals.org/Downloaded from
education in the United Kingdom than in Germany: re-
cent OECD figures show that 55 per cent of the individ-
uals at the typical graduation age graduate from ISCED
5A level (largely theoretically based tertiary pro-
grammes) in the United Kingdom, in contrast to only 30
per cent in Germany (OECD, 2013,p.56).
2
Due to
these differences in access and participation in HE, het-
erogeneity of tertiary students is lower in Germany than
in the United Kingdom, and the signalling capacity of
educational credentials is likely to be weaker in the
United Kingdom than in Germany.
Moreover, differentiation within HE differs between
the two countries: Germany has kept its binary system
of HE differentiating between university (offering more
academic curricula) and Fachhochschule (providing
more practical training). The United Kingdom moved
from a binary to a diversified system when polytechnics
gained university status in 1992. This has not ended
stratification in the UK HE sector, as a large differenti-
ation in prestige and student composition exists between
Russell Group universities, pre- and post-92 universities,
and further education colleges (Boliver, 2011;Iannelli,
Gamoran, and Paterson, 2011). Due to the variety of
secondary qualifications which allow entry into HE (e.g.
A-Levels/Highers, National Vocational Qualifications/
Scottish Vocational Qualifications, Business and
Technology Education Council (BTEC) qualifications,
or International Baccalaureate) and the lack of standar-
dized entry requirements into HE, admission criteria
strongly vary between HE institutions in the United
Kingdom (Cheung and Egerton, 2007). The more di-
verse institutional features in the United Kingdom may
affect the relation between social origin and labour
market returns in two ways. First, with regard to access
and choices within HE, the more diversified system in
the United Kingdom allows students from higher family
backgrounds to distinguish themselves from lower back-
grounds by choosing more prestigious fields of study
and institutions. Second, from the employer’s perspec-
tive, these less standardized features of the UK system
deliver less distinct and clear information to future em-
ployers who in turn may use other criteria, beyond the
degree, for selecting applicants such as the aforemen-
tioned criteria field of study and HE institution. In the
recruitment process, an applicant’s social origin might
additionally be used as a proxy for inherited social and
cultural capital, or employers may simply prefer gradu-
ates from higher classes if they perceive them to be more
alike to themselves (Kanter, 1977). In a field experi-
ment, Jackson (2009) shows that UK employers take
into account indicators of higher-class backgrounds and
discriminate in favour of offspring from higher classes.
Education–Labour Market Linkage:
Occupational versus Internal Labour
Market
Germany is usually described as the prototype of a la-
bour market that is occupationally segmented
(Marsden, 1990;Gangl, 2003). In occupational labour
markets, employers strongly rely on labour market en-
trants’ educational certificates in their hiring decisions.
The prevalence of occupationally segmented labour
markets in Germany is predominantly attributed to the
dual system of apprenticeship (Marsden, 1990).
However, strong linkages between HE and the graduate
labour market exist as well (Leuze, 2007). Entry into
professions and the high ranks of public sector is
strongly linked to university education (Klein, 2011).
For the traditional professions, such as medicine or law,
both universities and the state are responsible for profes-
sional training, and examinations and certification are
state-controlled.
The United Kingdom is often classified as a firm
internal labour market (e.g. Gangl, 2003). In such la-
bour markets, employers fill high-skilled vacancies
from the existing pool of employees, who have already
acquired firm-specific skills and expertise. The United
Kingdom has a less credentialist tradition than
Germany. The professions developed outside the univer-
sities and the state was completely absent in the organ-
ization and certification of modern professions (Abbott,
1988,p.202). Each profession developed a system of
self-training and self-certification which differed from
one professional association to another. Although pro-
fessional training is increasingly integrated into post-
graduate programmes (e.g. Master of Laws, Master of
Sciences), professional associations are still regarded as
responsible for professional development and training
outside academia (Leuze, 2007). Even for access to man-
agerial positions, educational certificates tend to be less
important in the UK than in the German labour market
(Stewart et al., 1994). Overall, linkages between HE and
occupational destinations are clearly less pronounced in
the United Kingdom than in Germany.
Hypotheses
Based on theories reviewed and country institutional dif-
ferences discussed in the previous sections, we hypothe-
size that:
(1) Social inequalities among graduates in occupa-
tional destinations are higher in the United
Kingdom than in Germany due to the lower social
selection in education, higher competition among
4European Sociological Review, 2015, Vol. 0, No. 0
at Edinburgh University on February 18, 2015http://esr.oxfordjournals.org/Downloaded from
graduates, and weaker links between education
and labour market (gross effect).
(2a) In both countries, social inequalities in occupa-
tional outcomes are mostly explained by qualita-
tive education differences.
However, the mediating role of qualitative dif-
ferences is expected to be larger in the United
Kingdom than in Germany due to the lower signal-
ling capacity of the educational level and the stron-
ger competition among graduates in the United
Kingdom. Students from higher social origins need
to differentiate themselves more strongly from
students from lower social origins in the United
Kingdom than in Germany.
(2b) Taking into account qualitative differences, the re-
maining net effect of social background is expected
to be larger in the United Kingdom than in
Germany. Social origin has a stronger impact on
job allocation, as UK employers need to distin-
guish between the greater and less socially selected
number of graduates and they have more scope to
decide which characteristics to regard as important
in the selection process.
(3) Country differences in the gross and net effect of so-
cial origin become smaller five years after gradu-
ation because graduates’ productivity is now
observable and further occupational destinations
are more likely to be acquired through graduates’
own resources and work experience.
Data and Methods
For our empirical analysis, we use harmonized cross-
sectional data from the ‘Research into Employment and
Professional Flexibility’ (REFLEX) survey.
3
This survey
interviewed individuals who graduated from tertiary in-
stitutions in 1999/2000 five years after their graduation
(2005/2006) in 16 European countries, including
Germany and the United Kingdom. It contains detailed
and comparable information about graduates’ social ori-
gin, their studies, the transition from school to work,
and current labour market situation. Graduates from
postsecondary vocationally oriented study programmes
(ISCED 5B) were excluded from the sampling
population.
We restrict our sample to individuals under the age
of 30 in the United Kingdom and 35 in Germany at the
time of graduation. For Germany, we consider a higher
age ceiling, as graduates tend to stay longer in HE than
UK graduates.
4
In addition, we exclude those individuals
who lived abroad at the time of the interview. The
number of cases analysed are 1,548 for Germany and
1,155 for the United Kingdom.
We use two different measures of occupational
position for the first stable job (i.e. the first job lasting
for >6 months) as well as the job occupied five years
after graduation. The International Socio-Economic
Index of Occupational Status (ISEI)(Ganzeboom, De
Graaf and Treiman, 1992) is a continuous measure of
occupational positions in terms of average earnings and
education. Additionally, we use the EGP class schema
(Erikson and Goldthorpe, 1992) and concentrate the
analysis on access to the higher-service class (EGP I) ver-
sus all other classes.
With regard to social origin, the REFLEX data do
not provide information on parents’ social class or occu-
pation. Thus, we have to rely on mothers’ and fathers’
highest educational attainment. We distinguish three
categories: both parents with an ISCED 1–4-level quali-
fication (below tertiary level), one parent with an ISCED
5–6-level qualification (tertiary programmes), and both
parents with an ISCED 5–6-level qualification (tertiary
programmes). We are likely to overestimate the effect of
parental education, as we lack measures of class origin
or parental status, while, at the same time, we underesti-
mate the full extent of social inequalities in labour mar-
ket returns among graduates (Bukodi and Goldthorpe,
2013).
We control for a range of potentially mediating
factors. Field of study is classified into nine categories:
Teaching, Medicine/Pharmacy, Law, Economics/
Business, Science, Technical subjects, Social sciences,
Humanities, and Welfare/Agriculture (e.g. social work
and counselling, forestry). The HE institution is
operationalized as a binary variable in Germany differ-
entiating between university and Fachhochschule. In the
United Kingdom, we distinguish between old univer-
sities, pre-92 and post-92 universities: the ‘old univer-
sities’ are the oldest and most prestigious universities,
holding high international reputation and attracting
most of the research funding; the ‘pre-1992 universities’
include all the universities which were created in the
1960s and before; and the ‘post-1992 universities’ are
composed of the ex-polytechnics, which in 1992 were
given the status of HE institutions, and other universities
founded after 1992.
5
Although tertiary graduates should be more homoge-
neous in terms of ability and motivation than secondary
school leavers, we control for the final grade upon
graduation. For the United Kingdom, this is operational-
ized by the class of degree which comprises the catego-
ries first class honours, upper second (2:1), lower second
(2:2), and third or pass. For Germany, we have
European Sociological Review, 2015, Vol. 0, No. 0 5
at Edinburgh University on February 18, 2015http://esr.oxfordjournals.org/Downloaded from
information on the average grade which is a continuous
measure ranging from 1.0 to 4.0. In order to make per-
formance comparable to the British undergraduate clas-
sification, we grouped the average grade as follows:
grades 1.0–1.5: first class; grades 1.6–2.0: second class,
upper division; grades 2.1–2.5: second class, lower div-
ision; grades 2.6–4.0: third or lower.
Because graduates from different social origins may
have different enrolment patterns into postgraduate pro-
grammes and a PhD is accompanied by specific labour
market outcomes, we differentiate between ‘being
enrolled in a PhD program or acquired a PhD’ and
‘acquired no PhD’ both at the stage of the first job and
five years after graduation. Finally, we control for age,
gender, and ethnic origin in all analyses.
6
Appendix Table A1 reports country differences in the
distributions of the relevant variables. The proportion of
graduates in the United Kingdom who originate from
parents with lower educational attainment is substan-
tially larger than in Germany. More than two thirds
of German graduates studied at university rather than
Fachhochschule. Almost half of the UK graduates were
enrolled in post-92 universities, while old and pre-92
universities are almost equally attended. German gradu-
ates more frequently studied technical disciplines, medi-
cine, and law. By contrast, UK graduates were more
often found in science, humanities, and social sciences.
Final grades are rather concentrated in the middle cate-
gories in the United Kingdom while grades in Germany
are distributed almost evenly across the grades’ range.
A larger share of graduates pursues a PhD in Germany
than in the United Kingdom.
Regarding labour market outcomes, a higher propor-
tion of graduates in Germany attain service class pos-
itions in their first job than in the United Kingdom.
Average ISEI is larger in Germany as well. While there
seem to be only few changes in terms of labour market
returns in Germany during the first five years after
graduation, there is a clear improvement in occupational
positions among HE graduates in the United Kingdom.
In our multivariate analyses, we conduct OLS regres-
sion models on ISEI-score for both countries separately.
In the analysis of the binary outcome ‘higher-service
class vs. all other classes’, we run linear probability
models with heteroskedasticity-robust standard errors
in order to compare social origin estimates across nested
models and between the two countries.
7
A comparison
of coefficients from non-linear probability models
across nested models or different subsamples would not
be appropriate, as these coefficients may be biased due
to differences in unobserved heterogeneity (Mood,
2010). To avoid an excessive reduction in the number of
cases analysed, we imputed missing values by multiple
imputation via ‘chained equations’ (Rubin, 1996) using
STATA module ‘ice’ (Royston, 2005).
Results
Occupational Position
Figure 1 shows the effects of parent’s education on
graduates’ first job and their job five years after gradu-
ation, both measured by the ISEI. Coefficients and 95%
confidence intervals are presented.
8
The first model in each figure (M1) shows the gross
effect of parental education after controlling for gender,
ethnic origin, and age. Comparing the estimates across
countries, we are able to evaluate the first hypothesis.
In the second model (M2) the information on ‘HE insti-
tution’ and in the third model (M3) ‘field of study’ are
included. The final model (M4) introduces graduates’
‘final grade’ and ‘PhD status’. Changes across models
(M1–M4) in both countries provide insights on hypoth-
esis 2a. In order to draw conclusions regarding hypoth-
esis 2b, we compare the net effect in the fourth model
across the United Kingdom and Germany. For evaluat-
ing the third hypothesis, we need to compare the associ-
ation between social origin and ISEI in the first job
(left-hand side) with the one 5 years after graduation
(right-hand side).
Looking at Figure 1, the first impression is of striking
similarities between the two countries. Focusing our
attention on the first job (the upper and lower graphs on
the left-hand side), the results show that parental educa-
tion significantly (and to a similar extent) affects gradu-
ates’ occupational status in both countries (M1). In the
United Kingdom and Germany, highly educated parents
(ISCED 5 þ6) are able to transmit a similar advantage
to their graduate children compared to lower educated
parents (ISCED 1–4). In Germany, graduates with me-
dium educated parents (one parent with ISCED 5 þ6)
have significant occupational advantages as well.
However, these advantages are of modest magnitude,
the occupational status of children of highly educated
parents being between 4 and 2 points higher (on a scale
ranging from 16 to 90) than children of lower educated
parents.
Introducing ‘type of HE institution’ slightly reduces
the effect of parental education in both countries (M2).
Most of the effect of parental education is mediated
by ‘field of study’ in both countries (M3) and high par-
ental education becomes insignificant. In the United
Kingdom, this effect decreases further when taking into
account ‘final grade’ and ‘PhD status’ (M4). In both
6European Sociological Review, 2015, Vol. 0, No. 0
at Edinburgh University on February 18, 2015http://esr.oxfordjournals.org/Downloaded from
countries parental education has no net effect on the oc-
cupational status of their children’s first job after taking
qualitative educational differences into account.
When looking at the influence of parental education
on graduates’ occupational status five years after gradu-
ation (the upper and lower graph on the right-hand
side), these effects are reduced in both countries and
they are mediated by HE institution (M2) and particu-
larly by field of study (M3). Overall, the impact of social
origin on the socio-economic status weakens throughout
the career in both countries.
In order to test whether country differences are sig-
nificant, we pooled the data from the two countries
and included interaction terms between country and
parental education.
9
Confirming the results presented in
Figure 1, no significant differences were found between
Germany and the United Kingdom when occupational
destinations were measured by the ISEI (see Appendix
Tables A2 and A3).
Access to the Higher-Service Class
Figure 2 presents the results of the analyses on the effect
of parental education on accessing the higher-service
class. The nested linear probability models follow the
same structure for estimation and layout of display as
in Figure 1.
10
This time the results show some important differ-
ences between Germany and the United Kingdom.
While both gross effects of high parental education on
service-class attainment are significant, the effect is
much stronger in the United Kingdom than in Germany
-5
-2.5
0
2.5
5
7.5
10
ISEI points
Model 1 Model 2 Model 3 Model 4
both parents with ISCED 5+6
one parent with ISCED 5+6
First job
-5
-2.5
0
2.5
5
7.5
10
ISEI points
Model 1 Model 2 Model 3 Model 4
both parents with ISCED 5+6
one parent with ISCED 5+6
Five years after grad.
Germany
-5
-2.5
0
2.5
5
7.5
10
ISEI points
Model 1 Model 2 Model 3 Model 4
both parents with ISCED 5+6
one parent with ISCED 5+6
First job
-5
-2.5
0
2.5
5
7.5
10
ISEI points
Model 1 Model 2 Model 3 Model 4
both parents with ISCED 5+6
one parent with ISCED 5+6
Five years after grad.
UK
Figure 1. Parental education effects on occupational destinations measured by ISEI in Germany and the United Kingdom
(OLS regressions)
Source: REFLEX, 2005, own calculations; Notes: N ¼1,548 for Germany and N¼1,155 for the United Kingdom; multiple imputation via chained equations;
ranges within capped spikes indicate 95% confidence intervals; M1 controls for gender, ethnic origin, and age; M2 ¼M1 þHE institution; M3¼M2 þfield
of study; M4¼M3 þfinal grade and ‘PhD status’. For the full set of estimates see Tables S1 and S2 in the Supplementary Material.
European Sociological Review, 2015, Vol. 0, No. 0 7
at Edinburgh University on February 18, 2015http://esr.oxfordjournals.org/Downloaded from
(M1). In the United Kingdom, graduates from a high
parental education background have a probability of at-
taining a service class that is 18 percentage points higher
than the probability of those with low educated parents.
The same figure in Germany is 10 percentage points.
The gross effect of medium educated parents is insignifi-
cant in both countries. When controlling for HE institu-
tion (M2), the effect of high parental education is only
slightly reduced in both countries, more so in the United
Kingdom than in Germany. When controlling for field
of study (M3), the effect of having ‘both parents with
ISCED 5 þ6’ becomes insignificant and almost cancels
out in Germany. In the United Kingdom, it becomes
smaller but remains substantially large and significant
at the 5 per cent level. This suggests that ‘field of study’
is a less important mediator for parental education ef-
fects in the United Kingdom than in Germany for entry
into higher-service class. The effect of high parental edu-
cation on accessing the higher-service class is somewhat
further reduced in the United Kingdom, when ‘final
grade’ and ‘PhD enrolment’ are included (M4).
However, in contrast to Germany, a net effect of high
parental education (10 percentage points and thus as
strong as the gross effect in Germany) on attaining top-
level occupations persists in the United Kingdom after
controlling for qualitative education differences.
Five years after graduation, the gross effect of
parental education on graduates’ chances of a higher-
service class position reduces in both countries (M1).
Although still stronger in the United Kingdom, parental
education effects become more similar. In Germany, the
effect of parental education becomes insignificant
when HE institution is taken into account (M2). Again,
‘field of study’ is a strong mediator for the association
-.1
-.05
0
.05
.1
.15
.2
.25
.3
Percentage points
Model 1 Model 2 Model 3 Model 4
both parents ISCED 5+6
one parent ISCED 5+6
First job
-.1
-.05
0
.05
.1
.15
.2
.25
.3
Percentage points
Model 1 Model 2 Model 3 Model 4
both parents ISCED 5+6
one parent ISCED 5+6
Five years after grad.
Germany
-.1
-.05
0
.05
.1
.15
.2
.25
.3
Percentage points
Model 1 Model 2 Model 3 Model 4
both parents ISCED 5+6
one parent ISCED 5+6
First job
-.1
-.05
0
.05
.1
.15
.2
.25
.3
Percentage points
Model 1 Model 2 Model 3 Model 4
both parents ISCED 5+6
one parent ISCED 5+6
Five years after grad.
UK
Figure 2. Parental education effects on accessing the higher-service class in Germany and the United Kingdom (linear probability
models)
Source: REFLEX, 2005, own calculations; Notes: N ¼1,548 for Germany and N¼1,155 for the United Kingdom; multiple imputation via chained equations;
ranges within capped spikes indicate 95% confidence intervals; M1 controls for gender, ethnic origin, and age; M2 ¼M1 þHE institution; M3¼M2 þfield
of study; M4¼M3 þfinal grade and ‘PhD status’. For the full set of estimates, see Tables S3 and S4 in the Supplementary Material.
8European Sociological Review, 2015, Vol. 0, No. 0
at Edinburgh University on February 18, 2015http://esr.oxfordjournals.org/Downloaded from
between parental education and higher-service class.
11
In contrast to the first job, no net effect of parental edu-
cation in the UK labour market prevails when graduates
move on in their working career. More meritocratic re-
cruitment or promotion criteria seem to have replaced
ascriptive forms of job allocation.
In our pooled analysis, country differences in the as-
sociation between high parental education and higher-
service class in the first job are confirmed in statistical
terms in the third model and are close to be significant
at conventional criteria in the fourth model (see
Appendix Table A4). Overall, country differences in
both gross (M1) and net effects (M4) of high parental
education are large (difference of 8 percentage points).
For the job five years after graduation, we do not see
any significant differences in the gross (M1) and net ef-
fect (M4) of parental education on higher-service class
attainment (see Appendix Table A5).
12
We run a fifth model (see Appendix Tables A3 and
A5) in the analysis on occupational position five years
after graduation to account for ISEI or higher-service
class in first job. Holding the occupational position
in the first job constant does not alter the conclusions
regarding parental education effects. However, the first
job is a much stronger predictor of the job five years
after graduation in Germany than in the United
Kingdom. This confirms that occupational destinations
are determined at labour market entry in Germany due
to much tighter links between education and the labour
market while graduates in the United Kingdom are more
likely to experience career advancement having started
from a lower level occupation. We conducted analyses
differentiated by gender to investigate whether these pat-
terns hold for both men and women (see Supplementary
Material, Tables S7–S10).
Overall, country differences as well as the role of medi-
ating factors are confirmed for both men and women.
Interestingly, we find that social origin effects on entry
into the higher-service class are larger among women than
men in both countries. However, this result should be con-
sidered with caution due to the low number of cases.
Discussion
This study aimed to provide a better understanding
of the relationship between social origin (measured by
parental education) and the transition from HE to the
labour market and of the role of national education
and labour market systems in shaping this relationship.
Our first research question asked whether, and the ex-
tent to which, social origin affects graduates’ occupa-
tional outcomes in Germany and the United Kingdom.
Given the different institutional settings in these two
countries, namely differences in selectivity and differen-
tiation of the education system at secondary and tertiary
levels as well as differences in the degree of education–
labour market linkage, we expected social differentials
to be greater in the United Kingdom than in Germany
(hypothesis 1). Our results only partly confirmed our
first hypothesis. When looking at the full range of occu-
pations that graduates entered (measured by the ISEI),
differences between social groups were small and no sig-
nificant country variation emerged. However, our first
hypothesis was confirmed when entry into the higher-
service class was analysed. Parental education had a
positive effect on this outcome in both Germany and the
United Kingdom, but its effect was stronger in the
United Kingdom (in particular in relation to having two
highly educated parents which would require further
investigation). We explained this result by the lower se-
lectivity of the HE system, the lower signalling power
of higher educational credentials, and their weaker links
with the labour market in the United Kingdom than in
Germany. These results also show that different meas-
ures of occupational outcomes can lead to different re-
sults. In our analysis, the use solely of the ISEI would
have concealed important national differences.
Our second research question asked about country
variation in the role of qualitative education differences
(measured by field of study and HE institution) in ex-
plaining the social origin effect on graduates’ occupa-
tional destinations. As expected we found that field of
study is an important mediator of social inequalities in
labour market outcomes in both countries. This finding
stresses the importance of analysing the role of ‘fields
of study’ in social mobility studies. However, contrary
to our expectations (hypothesis 2a), the mediating role
of field of study was found to be stronger in Germany
than in the United Kingdom. It is particularly evident
when entry into the higher-service class was analysed.
This may be explained by the stronger credentialist sys-
tem and the stronger linkage between education and la-
bour market in Germany than in the United Kingdom.
The labour market advantage of children from highly
educated parents in Germany is mostly explained by
their choice of more rewarding fields of study (such as
medicine and law).
In agreement with our hypothesis 2b, we found that
in the United Kingdom a net effect of parental education
on attaining a first job in the higher-service class persists
after controlling for qualitative education differences.
While in Germany social inequalities are transmitted
mainly via education (both vertically through early se-
lection and horizontally through field of study)
European Sociological Review, 2015, Vol. 0, No. 0 9
at Edinburgh University on February 18, 2015http://esr.oxfordjournals.org/Downloaded from
inequalities are only partly transmitted via education in
the United Kingdom. Social reproduction theories high-
light that the advantages of family of origin can be trans-
mitted through resources other than education, for
example cultural and social capital and ‘soft’ skills such
as communication skills, influencing skills, and personal
effectiveness. Many years ago Turner (1960) described
the United Kingdom as a ‘sponsored’ mobility system (in
contrast to the ‘contest’ mobility system of the United
States). Although as a whole, the United Kingdom is a
more fluid society compared to Germany (Breen and
Luijkx, 2004,p.72), our results show that the import-
ance of having a ‘sponsor’ to enter the top-level occupa-
tions may still be a feature of UK society (Tholen et al.,
2013). The ‘sponsor’ could be drawn from the family
social and occupational networks or simply embodied
in the educational institutions (e.g. private schools) at-
tended by socially advantaged children. Of course,
these resources can be more easily activated in a labour
market where jobs are less tightly linked to education
qualifications and less strongly regulated, such as in the
UK labour market.
Our data, however, suggest that more meritocratic
criteria are at work after the time of labour market
entry. In our third question, we asked whether country
differences in the effect of social origin persist five years
after graduation. The answer is negative: country differ-
ences were considerably smaller and not significant at
this later point in the occupational career. As expected
(hypotheses 3) the influence of social origin was found
to be stronger at labour market entry than five years
after graduation. Gross effects of parental education on
entry into the higher-service class still existed even five
years after graduation, but no direct effect remained
after taking into account qualitative differences between
graduates in both Germany and the United Kingdom.
Five years after graduation, the significant advantage
associated with highly educated parents could be ex-
plained by graduates’ field of study.
Our analysis was confined to studying the effect of
parental education on occupational destinations. Future
research could improve upon our study by using other
measures of social background characteristics (such as
social class of origin and parental income), by analysing
an array of labour market outcomes and a longer time
span in individuals’ occupational careers.
Notes
1 Different education systems have developed in the
United Kingdom over time. In the 1960s, the com-
prehensive re-organisation of secondary schools
was more widespread in Scotland and Wales than
in England. In England grammar schools have con-
tinued to exist and the private fee-paying sector has
always had a more conspicuous presence than in
Scotland. This has led to a different degree of differ-
entiation among schools in Scotland, England, and
Wales.
2 The graduation rates reported by the OECD for the
cohort analysed in this article, that is those grad-
uated in 1999/2000, are about 15% in Germany
and 35% in the United Kingdom (OECD, 2001,
p. 159).
3 For detailed information on the methodology of the
REFLEX survey, see the Supplementary Material
and Allen and Van der Velden (2007).
4 The typical graduation age from ISCED5A was 25
in Germany and 21 in the United Kingdom (OECD,
2002). In the Reflex data, 72% of UK graduates
were aged 20–24 on graduation while 64% of
German graduates were aged 25–29.
5 Information on HE institutions in the United
Kingdom is not included in the publicly available
REFLEX dataset. We thank Rolf Van der Velden
and Jim Allen for kindly providing us with this
aggregate measure for the United Kingdom.
6 In additional models we control for duration of study,
participation in a mandatory internship, or any other
work experience before and during graduates’ studies.
The results related to social origin effects did not
change with the inclusion of these activities.
7IntheSupplementary Material (Tables S5 and S6),
we also show estimates as average marginal effects
(AMEs) in our models on access to the higher-
service class. Regarding the association between so-
cial origin and top-level positions across nested mod-
els, we arrive at the same substantive conclusions.
8 For the full set of estimates, see Tables S1 and S2 in
the Supplementary Material.
9 Because HE institution is not directly comparable
between countries, we reduced the three UK catego-
ries to a binary variable differentiating between uni-
versities established before 1992 and the post-92
universities (similar to Fachhochschule).
10 For the full set of estimates, see Tables S3 and S4
in the Supplementary Material.
11 The changes in the estimates of ‘both parents with
ISCED 5 þ6’ on higher-service class (M1 to M4)
are significant at the 0.01% level at both labour
market stages in both countries.
12 As a robustness check we also recoded respond-
ents’ ISEI into a binary variable using the upper
decile of each country’s ISEI distribution vs. all
10 European Sociological Review, 2015, Vol. 0, No. 0
at Edinburgh University on February 18, 2015http://esr.oxfordjournals.org/Downloaded from
others. The results are similar to those for access
to higher-service class positions, confirming this
approach (results upon request).
Acknowledgements
We would like to thank the editor in chief and two anonymous
reviewers for valuable comments on an earlier version of this
article. The authors have also benefited from comments received
by members of the AQMeN (Applied Quantitative Methods
Network) ‘Education and Social Stratification’ strand (Emer
Smyth, Selina McCoy, Adam Gamoran, Lindsay Paterson,
Judith Offerhaus, Dafni Dima).
Funding
This work was conducted within the research centre AQMeN
(Applied Quantitative Methods Network) and supported by the
Economic and Social Research Council [grant number: ES/
K006460/1].
Supplementary Data
Supplementary data are available at ESR online.
References
Abbott, A. (1988). The System of Professions: An Essay on the
Division of Expert Labor. Chicago: The University of Chicago
Press.
Allen, J. and Van der Velden, R. (2007). The Flexible
Professional in the Knowledge Society: General Results of the
REFLEX Project. Research Centre for Education and the
Labour Market, Maastricht University.
Bol, T. and Van de Werfhorst, H. G. (2011). Signals and closure
by degrees: the education effect across 15 European
Countries. Research in Social Stratification and Mobility, 29,
119–132.
Boliver, V. (2011). Expansion, differentiation, and the persist-
ence of social class inequalities in British higher education.
Higher Education, 61, 229–242.
Bourdieu, P. (1986). The forms of capital. In Richardson, J. G.
(Ed.), Handbook for Theory and Research for the Sociology
of Education. New York: Greenwood Press, pp. 241–258.
Breen, R. (2004). Social Mobility in Europe. Oxford: Oxford
University Press.
Breen, R. (2010). Educational expansion and social mobility in
the 20th century. Social Forces, 89, 365–388.
Breen, R. and Jonsson, J. O. (2008). Explaining change in social
fluidity: educational equalization and educational expansion
in twentieth-century Sweden. American Journal of Sociology,
112, 1775–1810.
Breen, R. and Luijkx, R. (2004). Social mobility in Europe
between 1970 and 2000. In Breen, R. (Ed.), Social Mobility in
Europe. Oxford: Oxford University Press, pp. 37–75.
Bukodi, E. and Goldthorpe, J. H. (2013). Decomposing ‘Social
Origins’: the effects of parents’ class, status and education on
the educational attainment of their children. European
Sociological Review, 29, 1024–1039.
Cheung, S. Y. and Egerton, M. (2007). Great Britain: higher
education expansion and reform - changing educational
inequalities. In Shavit, Y., Arum, R., and Gamoran, A. (Eds.),
Stratification in Higher Education. A Comparative Study.
Stanford: Stanford University, pp. 195–219.
Coleman, J. S. (1988). Social capital in the creation of human
capital. The American Journal of Sociology, 94, 95–120.
Collins, R. (1979). The Credential Society: An Historical Sociology
of Education and Stratification. New York: Academic Press.
Erikson, R. and Goldthorpe, J. H. (1992). The Constant Flux.
A Study of Class Mobility in Industrial Societies. Oxford:
Clarendon Press.
Gangl, M. (2003). The structure of labour market entry in
europe: a typological analysis. In Mu¨ ller, W. and Gangl, M.
(Eds.), Transitions from Education to Work in Europe: The
Integration of Youth into EU Labour Markets. Oxford:
Oxford University Press, pp. 107–128.
Ganzeboom, H.B.G., De Graaf, P. M. and Treiman, D. J.
(1992). A standard international socio-economic index of
occupational status. Social Science Research, 21, 1–56.
Hout, M. (1988). More universalism, less structural mobility:
the American occupational structure in the 1980s. American
Journal of Sociology, 93, 1358–1400.
Iannelli, C., Gamoran, A. and Paterson, L. (2011). Scottish
Higher Education, 1987-2001: expansion through diversion.
Oxford Review of Education, 37, 717–741.
Iannelli, C. and Paterson, L. (2007). Education and social mobil-
ity in Scotland. Research in Social Stratification and Mobility,
25, 219–232.
Jackson, M. (2009). Disadvantaged through discrimination?
The role of employers in social stratification. British Journal
of Sociology,60, 669–691.
Jackson, M., Goldthorpe, J. H. and Mills, C. (2005). Education,
employers and class mobility. Research in Social Stratification
and Mobility, 23, 3–33.
Jacob, M., Gerth, M. and Weiss, F. (2014). Freshmens, sopho-
mores and finalists employment: social inequalities in student
jobs and their development during higher education. Paper
presented at the RC28 Spring Meeting, Budapest.
Jacob, M. and Weiss, F. (2011). From higher education to work.
Patterns of labor market entry in Germany and the US. Higher
Education, 60, 529–542.
Kanter, R. M. (1977). Some effects of proportions on group life:
skewed sex ratios and responses to token women. American
Journal of Sociology, 82, 965–990.
Klein, M. (2011). Trends in the association between educational
attainment and class destinations in West Germany: Looking
inside the service class. Research in Social Stratification and
Mobility, 29, 427–444.
Leuze, K. (2007). What makes for a good start? Consequences
of occupation-specific higher education for career mobility.
International Journal of Sociology, 37, 29–53.
European Sociological Review, 2015, Vol. 0, No. 0 11
at Edinburgh University on February 18, 2015http://esr.oxfordjournals.org/Downloaded from
Lucas, S. R. (2001). Effectively maintained inequality: education
transitions, track mobility, and social background effects.
American Journal of Sociology, 106, 1642–1690.
Mare, R. D. (1980). Social background and school continuation
decisions. Journal of the American Statistical Association, 75,
295–305.
Marsden, D. (1990). Institutions and labour mobility: occupa-
tional and internal labour markets in Britain, France, Italy and
West Germany. In Brunetta, R. and Dell’Aringa, C. (Eds.),
Labour Relations and Economic Performance. Venedig: The
Macmillan Press, pp. 414–438.
Mastekaasa, A. (2011). Social origins and labour market success
- stability and change over Norwegian birth cohorts 1950-
1969, European Sociological Review, 27, 1–15.
Mood, C. (2010). Logistic regression: why we cannot do what
we think we can do, and what we can do about it. European
Sociological Review, 26, 67–82.
OECD. (2001). Education at a Glance. Paris: OECD.
OECD. (2002). Education at a Glance. Paris: OECD.
OECD. (2013). Education at a Glance. Paris: OECD.
Reimer, D. and Pollak, R. (2010). Educational expansion and its
consequences for vertical and horizontal inequalities in access
to higher education in West Germany. European Sociological
Review, 26, 415–430.
Royston, P. (2005). Multiple imputation of missing values.
Update of Ice. The Stata Journal, 5, 527–536.
Rubin, D. B. (1996). Multiple imputation after 18 þyears.
Journal of the American Statistical Association, 91, 473–489.
Scherer, S. (2001). Early career patterns: a comparison of great
Britain and West Germany. European Sociological Review,
17, 119–144.
Schindler, S. (2014). Wege zur Studienberechtigung - Wege ins
Studium? Eine Analyse sozialer Inklusions- und
Ablenkungsprozesse. Wiesbaden: Springer VS.
Schindler, S. and Reimer, D. (2011). Differentiation and social
selectivity in German higher education. Higher Education, 61,
261–275.
Spence, M. (1973). Job market signaling. The Quarterly Journal
of Economics, 87, 355–374.
Stewart, R., Barsoux, J. L., Kieser, A., Ganter, H. D. and
Walgenbach, P. (1994) Managing in Britain and Germany.
New York: St. Martin’s Press.
Stiglitz, J. E. (1975). The theory of screening, education, and the
distribution of income. American Economic Review, 115,
283–299.
Tholen, G., Brown, P., Power, S. and Allouch, A. (2013). The
role of networks and connections in educational elites’ labour
market entrance. Research in Social Stratification and
Mobility, 34, 142–154.
Triventi, M. (2013). The role of higher education stratifica-
tion in the reproduction of social inequality in the labor
market. Research in Social Stratification and Mobility, 32,
45–63.
Turner, R. H. (1960). Sponsored and contest mobility
and the school system. American Sociological Review, 25,
855–867.
Van de Werfhorst, H. G., Sullivan, A. and Cheung, S. Y. (2003).
Social class, ability and choice of subject in secondary and ter-
tiary education in Britain. British Educational Research
Journal, 29, 41–62.
12 European Sociological Review, 2015, Vol. 0, No. 0
at Edinburgh University on February 18, 2015http://esr.oxfordjournals.org/Downloaded from
Appendix
Table A1. Descriptive characteristics of graduates
Germany United Kingdom
Mean SD Mean SD
Parental education
Both parents with ISCED 1–4 0.32 0.55
One parent with ISCED 5 þ6 0.36 0.23
Both parents with ISCED 5 þ6 0.32 0.22
Age at time of the survey 32.58 2.62 27.61 1.77
Gender (Ref. Male)
Female 0.49 0.60
Ethnic origin (Ref. Native)
Foreign 0.07 0.15
HE institution
University/Old university 0.69 0.28
Pre-92 university 0.24
Fachhochschule/Post-92 university 0.31 0.48
Field of study
Teaching
a
0.15 –
Medicine/Pharmacy 0.08 0.05
Law 0.07 0.03
Economics/Business 0.13 0.13
Science 0.09 0.24
Technical disciplines 0.23 0.07
Social sciences 0.04 0.13
Humanities 0.12 0.26
Welfare and social services 0.09 0.10
Class of Degree
Master 0.05
First class (1st) 0.21 0.08
Second class, upper division (2:1) 0.31 0.49
Second class, lower division (2:2) 0.25 0.29
Third (3rd) or ordinary (pass) 0.23 0.09
PhD enrolment/acquisition 0.15 0.04
Labour market outcomes
First job
Service class I 0.31 0.25
Service class II 0.46 0.37
Class III-VII 0.23 0.38
ISEI 67.48 11.73 57.54 20.90
Job five years after graduation
Service class I 0.32 0.31
Service class II 0.48 0.48
Class III-VII 0.20 0.21
ISEI 67.92 10.84 61.56 14.91
a
Graduates from teaching training programmes in Germany could be identified thanks to the availability of further informa-
tion on their teaching certificate (Lehramt). Unfortunately, such information is not available for the UK and future teachers are
categorised according to their subject of specialisation.
Source: REFLEX, 2005, own calculations.
Notes: N¼1,548 for Germany and N¼1,155 for the United Kingdom; multiple imputation via chained equations. Survey
weights were used to correct for over- or under-representation of certain levels or fields of higher education according to popula-
tion figures.
European Sociological Review, 2015, Vol. 0, No. 0 13
at Edinburgh University on February 18, 2015http://esr.oxfordjournals.org/Downloaded from
Table A2. Country differences in parental education effects on occupational destinations measured by ISEI in first job (OLS
regressions)
Model 1 Model 2 Model 3 Model 4
Country (Ref. Germany)
United Kingdom 44.95*** (9.99) 48.06*** (10.28) 20.43* (9.48) 15.36 (9.77)
Parental education (Ref.: Both parents ISCED 1–4)
Both parents ISCED 5 þ6 3.64*** (0.78) 2.92*** (0.77) 1.24 (0.65) 1.11 (0.65)
One parent ISCED 5 þ6 2.10** (0.72) 1.75* (0.72) 1.25* (0.62) 1.16 (0.62)
UK * both par. ISCED 5 þ60.41 (1.48) 0.05 (1.48) 0.00 (1.35) 0.57 (1.33)
UK * one par. ISCED 5 þ60.16 (2.04) 0.05 (2.10) 0.29 (1.96) 0.31 (1.94)
HE institution (Ref.: University)
FH/Post-92 university 3.56*** (0.62) 0.47 (0.72) 0.12 (0.74)
UK * FH/Post-92 1.65 (1.58) 0.49 (1.70) 1.15 (1.75)
Field of study (Ref.: Science)
Teaching
a
3.01** (1.12) 3.93*** (1.14)
Medicine/Pharmacy 18.48*** (1.15) 18.77*** (1.18)
Law 16.83*** (1.42) 18.12*** (1.54)
Economics/Business 2.94* (1.27) 2.03 (1.28)
Technical disciplines 3.87*** (1.13) 4.58*** (1.15)
Social science 0.47 (1.56) 0.65 (1.54)
Humanities 3.96** (1.32) 3.50** (1.33)
Welfare and social services 0.55 (1.37) 0.17 (1.34)
UK * med. 4.55* (2.10) 6.29* (2.45)
UK * law 14.61*** (2.82) 14.78*** (2.80)
UK * econ. 2.50 (1.98) 3.04 (1.97)
UK * tech. 2.94 (2.63) 3.92 (2.63)
UK * soc. 7.49*** (2.19) 6.78** (2.16)
UK * hum. 0.09 (2.12) 0.12 (2.06)
UK * welf. 8.23*** (2.33) 7.77** (2.31)
Grade (Ref.: 3rd or less)
First class (1st) 1.59 (0.87)
Second class, (2:1) 1.28 (0.78)
Second class, (2:2) 0.35 (0.83)
UK: Master
b
6.57** (2.20)
UK * first c. 0.98 (1.90)
UK * second c., (2:1) 0.11 (1.98)
UK * second c., (2:2) 7.45** (2.50)
PhD enrol./acquis. (yes ¼1) 1.77* (0.83)
UK * PhD 1.22 (2.76)
Constant 82.73*** (4.05) 84.85*** (4.03) 74.21*** (3.79) 71.67*** (3.94)
R
2
0.11 0.12 0.25 0.26
a
Graduates from teaching training programmes in Germany could be identified thanks to the availability of further information on their teaching certificate
(Lehramt). Unfortunately, such information is not available for the United Kingdom, and future teachers are categorized according to their subject of specialization.
b
A small number of graduates have a Master’s degree in the United Kingdom. In Germany, however, all graduates have a degree that is equivalent to a Master’s
degree (e.g. Diplom).
Source: REFLEX, 2005, own calculations.
Notes: Pooled model with N¼2,703; multiple imputation via chained equations; controlling for gender, ethnic origin, age, and respective interactions with coun-
try; standard errors in parentheses.
*p<0.05, **p<0.01, ***p<0.001.
14 European Sociological Review, 2015, Vol. 0, No. 0
at Edinburgh University on February 18, 2015http://esr.oxfordjournals.org/Downloaded from
Table A3. Country differences in parental education effects on occupational destinations measured by ISEI in job five years
after graduation (OLS regressions)
Model 1 Model 2 Model 3 Model 4 Model 5
Country (Ref. Germany)
United Kingdom 30.81*** (8.96) 33.94*** (9.02) 9.76 (8.38) 7.54 (8.62) 7.40 (7.58)
Parental education (Ref.: Both parents ISCED 1–4)
Both parents ISCED 5 þ6 2.40*** (0.71) 1.64* (0.71) 0.01 (0.61) 0.09 (0.61) 0.71 (0.48)
One parent ISCED 5 þ6 1.55* (0.64) 1.19 (0.63) 0.66 (0.55) 0.56 (0.55) 0.09 (0.43)
UK * both par. ISCED 5 þ60.25 (1.24) 0.18 (1.24) 0.45 (1.12) 0.22 (1.13) 0.60 (1.01)
UK * one par. ISCED 5 þ60.91 (1.39) 0.67 (1.40) 0.41 (1.29) 0.34 (1.29) 0.35 (0.94)
HE institution (Ref.: university)
FH/Post-92 university 3.74*** (0.56) 1.25 (0.64) 0.82 (0.65) 0.75 (0.48)
UK * FH/Post-92 2.00 (1.11) 1.37 (1.16) 1.50 (1.19) 0.98 (0.86)
Field of study (Ref.: Science)
Teaching
a
3.71** (1.11) 4.50*** (1.10) 2.31* (0.92)
Medicine/Pharmacy 18.86*** (1.27) 18.51*** (1.26) 8.05*** (1.16)
Law 15.78*** (1.57) 16.19*** (1.62) 6.10*** (1.41)
Economics/Business 0.59 (1.23) 0.03 (1.22) 1.16 (0.91)
Technical disciplines 4.93*** (1.08) 5.44*** (1.08) 2.88*** (0.81)
Social science 1.49 (1.45) 1.73 (1.42) 1.37 (1.18)
Humanities 3.06* (1.36) 2.63 (1.34) 0.68 (1.07)
Welfare and social services 0.56 (1.31) 0.89 (1.28) 0.98 (0.93)
UK * med. 1.26 (1.84) 2.47 (2.04) 1.80 (1.86)
UK * law 8.53** (2.85) 8.17** (2.85) 0.44 (2.61)
UK * econ. 4.54** (1.74) 4.83** (1.75) 3.71* (1.41)
UK * tech. 5.39* (2.12) 5.80** (2.15) 3.54* (1.61)
UK * soc. 7.39*** (2.04) 7.13*** (2.01) 4.04* (1.77)
UK * hum. 0.25 (1.83) 0.05 (1.83) 0.29 (1.47)
UK * welf. 8.68*** (2.12) 8.58*** (2.12) 5.15** (1.71)
Grade (Ref.: 3rd or less)
First class (1st) 0.37 (0.76) 0.51 (0.58)
Second class, (2:1) 0.24 (0.69) 0.47 (0.54)
Second class, (2:2) 0.82 (0.72) 0.62 (0.56)
UK: Master
b
2.51 (1.96) 0.22 (1.64)
UK * first c. 1.19 (1.67) 0.90 (1.39)
UK * second c., (2:1) 0.36 (1.76) 0.35 (1.47)
UK * second c., (2:2) 2.09 (2.09) 1.22 (1.80)
PhD enrol./acquis. (yes ¼1) 2.26** (0.74) 1.27 (0.63)
UK * PhD 1.63 (2.37) 1.30 (2.05)
ISEI in first job 0.56*** (0.03)
UK * ISEI in first job 0.11* (0.05)
Constant 79.20*** (3.71) 81.43*** (3.68) 70.54*** (3.48) 69.11*** (3.56) 29.17*** (3.47)
R
2
0.08 0.09 0.28 0.29 0.56
a
Graduates from teaching training programmes in Germany could be identified thanks to the availability of further information on their teaching certificate
(Lehramt). Unfortunately, such information is not available for the United Kingdom, and future teachers are categorized according to their subject of specialization.
b
A small number of graduates have a Master’s degree in the United Kingdom. In Germany, however, all graduates have a degree that is equivalent to a Master’s
degree (e.g. Diplom). Therefore, ‘having a Master’s degree’ is included only in the United Kingdom as a further measure of academic performance.
Source: REFLEX, 2005, own calculations.
Notes: Pooled model with N¼2,703; multiple imputation via chained equations; all models control for gender, ethnic origin, age, and respective interactions with
country; standard errors in parentheses.
*p<0.05, **p<0.01, ***p<0.001.
European Sociological Review, 2015, Vol. 0, No. 0 15
at Edinburgh University on February 18, 2015http://esr.oxfordjournals.org/Downloaded from
Table A4. Country differences in parental education effects on accessing the higher-service class in first job (linear prob-
ability models)
Model 1 Model 2 Model 3 Model 4
Country (Ref. Germany)
United Kingdom 1.17*** (0.27) 1.22*** (0.27) 0.45 (0.24) 0.48 (0.25)
Parental educ. (Ref.: Both parents ISCED 1–4)
Both parents ISCED 5 þ6 0.10** (0.03) 0.09** (0.03) 0.01 (0.03) 0.01 (0.03)
One parent ISCED 5 þ6 0.05 (0.03) 0.05 (0.03) 0.02 (0.02) 0.02 (0.02)
UK * both par. ISCED 5 þ60.08 (0.05) 0.07 (0.05) 0.10* (0.04) 0.08 (0.04)
UK * one par. ISCED 5 þ60.01 (0.04) 0.02 (0.04) 0.02 (0.04) 0.01 (0.04)
HE institution (Ref.: university)
FH/Post-92 university 0.02 (0.03) 0.02 (0.03) 0.05 (0.03)
UK * FH/Post-92 0.07 (0.04) 0.03 (0.04) 0.02 (0.04)
Field of study (Ref.: Science)
Teaching
a
0.17*** (0.04) 0.21*** (0.04)
Medicine/Pharmacy 0.69*** (0.05) 0.75*** (0.05)
Law 0.63*** (0.06) 0.62*** (0.06)
Economics/Business 0.09 (0.05) 0.07 (0.05)
Technical disciplines 0.22*** (0.05) 0.21*** (0.05)
Social science 0.11 (0.07) 0.09 (0.07)
Humanities 0.13** (0.04) 0.15*** (0.04)
Welfare and social services 0.00 (0.05) 0.01 (0.05)
UK * med. 0.10 (0.07) 0.10 (0.07)
UK * law 0.66*** (0.10) 0.62*** (0.10)
UK * econ. 0.18** (0.07) 0.14* (0.07)
UK * tech. 0.13 (0.08) 0.11 (0.08)
UK * soc. 0.26** (0.08) 0.21** (0.08)
UK * hum. 0.08 (0.06) 0.03 (0.06)
UK * welf. 0.19** (0.07) 0.14* (0.07)
Grade (Ref.: 3rd or less)
First class (1st) 0.03 (0.03)
Second class, (2:1) 0.00 (0.03)
Second class, (2:2) 0.01 (0.03)
UK: Master
b
0.21** (0.07)
UK * first c. 0.07 (0.05)
UK * second c., (2:1) 0.00 (0.05)
UK * second c., (2:2) 0.16* (0.08)
PhD enrol./acquis. (yes ¼1) 0.17*** (0.03)
UK * PhD 0.37*** (0.09)
Constant 0.56*** (0.15) 0.58*** (0.15) 0.28* (0.13) 0.36** (0.14)
R
2
0.03 0.04 0.27 0.29
a
Graduates from teaching training programmes in Germany could be identified thanks to the availability of further information on their teaching certificate
(Lehramt). Unfortunately, such information is not available for the United Kingdom, and future teachers are categorized according to their subject of specialization.
b
A small number of graduates have a Master’s degree in the United Kingdom. In Germany, however, all graduates have a degree that is equivalent to a Master’s de-
gree (e.g. Diplom). Therefore, ‘having a Master’s degree’ is included only in the United Kingdom as a further measure of academic performance.
Source: REFLEX, 2005, own calculations.
Notes: Pooled model with N¼2,703; multiple imputation via chained equations; all models control for gender, ethnic origin, age, and respective interactions with
country; standard errors in parentheses.
*p<0.05, **p<0.01, ***p<0.001.
16 European Sociological Review, 2015, Vol. 0, No. 0
at Edinburgh University on February 18, 2015http://esr.oxfordjournals.org/Downloaded from
Table A5. Country differences in parental education effects on accessing the higher-service class in job five years after
graduation (linear probability models)
Model 1 Model 2 Model 3 Model 4 Model 5
Country (Ref. Germany)
United Kingdom 0.89** (0.28) 0.96*** (0.28) 0.37 (0.26) 0.50 (0.27) 0.21 (0.25)
Parental education (Ref.: Both parents ISCED 1–4)
Both parents ISCED 5 þ6 0.08* (0.03) 0.06 (0.03) 0.02 (0.03) 0.01 (0.03) 0.02 (0.02)
One parent ISCED 5 þ6 0.04 (0.03) 0.04 (0.03) 0.01 (0.03) 0.01 (0.03) 0.00 (0.02)
UK * both par. ISCED 5 þ60.06 (0.05) 0.06 (0.05) 0.09 (0.05) 0.07 (0.05) 0.03 (0.04)
UK * one par. ISCED 5 þ60.05 (0.04) 0.05 (0.04) 0.01 (0.04) 0.02 (0.04) 0.02 (0.03)
HE institution (Ref.: University)
FH/Post-92 university 0.06* (0.03) 0.04 (0.03) 0.06 (0.03) 0.03 (0.03)
UK * FH/Post-92 0.02 (0.04) 0.01 (0.04) 0.03 (0.04) 0.01 (0.04)
Field of study (Ref.: Science)
Teaching
a
0.18*** (0.04) 0.22*** (0.04) 0.09** (0.03)
Medicine/Pharmacy 0.68*** (0.05) 0.70*** (0.05) 0.22*** (0.05)
Law 0.59*** (0.06) 0.54*** (0.06) 0.15* (0.06)
Economics/Business 0.09 (0.05) 0.05 (0.05) 0.01 (0.04)
Technical disciplines 0.19*** (0.05) 0.16** (0.05) 0.02 (0.04)
Social science 0.18* (0.08) 0.16* (0.07) 0.10 (0.06)
Humanities 0.10* (0.05) 0.13** (0.05) 0.03 (0.04)
Welfare and social services 0.02 (0.05) 0.05 (0.05) 0.04 (0.04)
UK * med. 0.10 (0.07) 0.06 (0.07) 0.11 (0.07)
UK * law 0.40*** (0.10) 0.32** (0.11) 0.08 (0.10)
UK * econ. 0.10 (0.07) 0.05 (0.07) 0.04 (0.06)
UK * tech. 0.19* (0.08) 0.15 (0.09) 0.06 (0.07)
UK * soc. 0.24** (0.09) 0.20* (0.09) 0.08 (0.08)
UK * hum. 0.09 (0.06) 0.05 (0.06) 0.06 (0.05)
UK * welf. 0.10 (0.07) 0.05 (0.08) 0.02 (0.06)
Grade (Ref.: 3rd or less)
First class (1st) 0.05 (0.03) 0.07* (0.03)
Second class, (2:1) 0.03 (0.03) 0.03 (0.02)
Second class, (2:2) 0.02 (0.03) 0.02 (0.03)
UK: Master
b
0.19* (0.08) 0.09 (0.06)
UK * first c. 0.10 (0.06) 0.07 (0.05)
UK * second c., (2:1) 0.03 (0.06) 0.03 (0.05)
UK * second c., (2:2) 0.09 (0.08) 0.01 (0.07)
PhD enrol./acquis. (yes ¼1) 0.13*** (0.03) 0.02 (0.03)
UK * Phd 0.34*** (0.09) 0.14 (0.09)
Higher-service class in first job (yes ¼1) 0.64*** (0.03)
UK * higher-service class in first job 0.16** (0.05)
Constant 0.60*** (0.15) 0.64*** (0.15) 0.40** (0.14) 0.53*** (0.14) 0.29* (0.11)
R
2
0.02 0.03 0.23 0.24 0.47
a
Graduates from teaching training programmes in Germany could be identified thanks to the availability of further information on their teaching certificate
(Lehramt). Unfortunately, such information is not available for the United Kingdom, and future teachers are categorized according to their subject of specialization.
b
A small number of graduates have a Master’s degree in the United Kingdom. In Germany, however, all graduates have a degree that is equivalent to a Master’s
degree (e.g. Diplom).
Source: REFLEX, 2005, own calculations.
Notes: Pooled model with N¼2,703; multiple imputation via chained equations; all models control for gender, ethnic origin, age, and respective interactions with
country; standard errors in parentheses.
*p<0.05, **p<0.01, ***p<0.001.
European Sociological Review, 2015, Vol. 0, No. 0 17
at Edinburgh University on February 18, 2015http://esr.oxfordjournals.org/Downloaded from
