Circular job-related spatial mobility in Germany:
Comparative analyses of two representative surveys on the 
forms, prevalence and relevance in the context of partnership 
and family development 

Heiko Rüger, Michael Feldhaus, Katharina S. Becker, Monika Schlegel

Abstract: Over the past few decades, employees have had to come to terms with 
increased demands of the labour market requiring greater fl exibility and mobility. 
At the same time, increasingly versatile and complex forms of job-related spatial 
mobility are emerging. Consequently, the correlation between job mobility patterns 
and family-related processes is attracting more and more attention in the fi eld of 
mobility and family research. However, to date there has rarely been a standard by 
which to systematically record and analyse job mobility. “Job Mobilities and Family 
Lives in Europe” (JobMob), a comparative European survey, and the “Panel Analy-
sis of Intimate Relationships and Family Dynamics” (pairfam) constitute two sets of 
representative data for Germany, which provide largely comparable operationalisa-
tions for several forms of circular job mobility, thus allowing us to systematically 
perform comparative analyses. For the fi rst time ever in this fi eld of research, it is 
now possible to subject fi ndings to a direct reciprocal validation process and to 
check whether general rules and correlations can be derived from them. 

In this regard, the present article aims at achieving three essential objectives. 
First, we will introduce a common indicator for circular job mobility patterns found 
in the two surveys. On the basis of this common indicator, we will comparatively 
analyse the prevalence of different mobility forms and their composition accord-
ing to key socio-demographic characteristics. In addition, we will use multivariate 
analyses to illustrate the relevance of job mobility for partnership and family devel-
opment. Results suggest mobility patterns to be an important individual context 
factor when explaining processes relevant to partnerships and family. In particular, 
women who exhibit some degree of job mobility are less often married and rarely 
have children.

Keywords: Circular job-related spatial mobility · Job commuting · Business travel · 
Institutionalisation of relationships · Childlessness

Comparative Population Studies – Zeitschrift für Bevölkerungswissenschaft
Vol. 36, 1 (2011): 221-248 (Date of release: 26.01.2012)

© Federal Institute for Population Research 2012  URL: www.comparativepopulationstudies.de
DOI: 10.4232/10.CPoS-2011-05en   URN: urn:nbn:de:bib-cpos-2011-05en7



•    Heiko Rüger, Michael Feldhaus, Katharina S. Becker, Monika Schlegel222

1 Introduction 

There is a long tradition of research on job mobility in the fi eld of sociology and in 
life course research. Classic examples include the early biographical study of Tho-
mas and Znaniecki (1918-1920) on the immigration of Polish peasants to European 
countries and America. Especially in the past, labour migration has been the sub-
ject matter of a variety of studies. On the one hand, empirical data shows that the 
prevalence of residential mobility (international and internal migration) is relatively 
low; at least it is lower than analysts of the current globalisation processes have 
assumed or than it could be expected on the basis of commonly expressed state-
ments by members of this increasingly “mobile” society (Sennett 1998; Castells 
2003; Urry 2007). However, circular forms of job mobility – such as long-distance or 
weekend commuting or frequent business trips – have gained empirical relevance 
in recent decades (Schneider et al. 2002a/b; Schneider et al. 2008; Schneider et al. 
2009; Haas/Hamann 2008; Ruppenthal/Lück 2009). In thus far, the theory advanced 
by Zelinsky as in the early 1970s has been confi rmed. His theory holds that in the 
late modern stages of mobility, more complex forms of mobility will emerge, and 
that, as a consequence, “circulatory movements” will gain signifi cance (Zelinsky 
1971: 245). 

Recent research has more strongly focused on this development; with the help of 
recent surveys, the amount of available data has improved signifi cantly (Lück/Sch-
neider 2010; Schneider et al. 2002a; Schneider/Meil 2008; Schneider/Collet 2010; 
Statistisches Bundesamt 2005). In the context of quantitative life course research 
that focuses on partnership and family development processes, the two surveys 
“Job Mobilities and Family Lives in Europe” (JobMob) (Schneider/Meil 2008) and the 
“Panel Analysis of Intimate Relationships and Family Dynamics” (pairfam) (Huinink 
et al. 2011) are particularly noteworthy. 

Job mobility, and even circular job mobility, may in fact exert a crucial infl uence 
on a couple’s or a family’s everyday life. Overall, its scale, the amount of time re-
quired for it, the fi nancial expense involved (e.g. Rüger/Ruppenthal 2010) and the 
psychological and physical consequences that may result from circular job mobility 
(e.g. Schneider et al. 2009b; Limmer/Rüger 2010) potentially constitute relevant pa-
rameters of the opportunity structure affecting partnerships and families (Schnei-
der 2005; Limmer/Schneider 2008; Huinink/Feldhaus 2008).

Even though research in this area is intensifying, the potential of existing stud-
ies has not been fully exhausted. Due to the fact that the two mentioned studies 
on spatial job mobility have been guided by an infl uential study by Schneider et al. 
(2002a/b), they can operationalise a wide range of different forms of mobility in the 
same systematic way. For the fi rst time in Germany, it is now possible to conduct 
comparative analyses on the topic of mobility and family research on the basis of 
two representative surveys. By taking advantage of reciprocal validation, one’s abil-
ity to generalise the fi ndings in this fi eld of research will take a major step forward 
(e.g. Raithel 2008). 

In this regard, the present article aims at achieving three essential objectives. 
First, we will present a common indicator for the circular job mobility patterns found 



Circular job-related spatial mobility in Germany    • 223

in the two surveys. On the basis of this common indicator, we will conduct a com-
parative analysis of the prevalence of a variety of forms of mobility and their com-
position according to key socio-demographic characteristics for both samplings. 
Due to the fact that both studies focus on issues relating to partnerships and family 
structures, we will use multivariate analyses to illustrate the relevance of circular job 
mobility for the development of relationship as well as family structures. 

2 Data and operationalisations

2.1 Data recording circular job-related spatial mobility in Germany

As concerns job mobility patterns, a key differentiation is made between residential 
and circular forms of mobility (e.g. Limmer/Schneider 2008). While mobility research 
in the past on the processes affecting partnership and family has concentrated on 
analysing residential forms of mobility, such as internal and international migration 
(e.g. Wagner 1989; Huinink/Wagner 1989; Kley 2009; Kulu 2005), this article focuses 
exclusively on circular forms of mobility, such as business travel, weekend commut-
ing, and other mobile forms of working. 

For the purpose of investigating the correlation between circular job mobility 
and related processes that are relevant to life course research, such as partnership 
and family developments and lifestyle choices, the study conducted by Norbert F. 
Schneider and his colleagues was ground-breaking for Germany (Schneider et al. 
2002a/b). It also constituted an essential fi rst step for the European study entitled 
”Job Mobilities and Family Lives in Europe” (Schneider/Meil 2008). Even though 
several other studies address the commuting behaviour of the working population,1 
for instance, those studies did not gather any data on partnership and family dynam-
ics. Oftentimes, they are more likely to focus on socio-professional, health-related, 
geographic or infrastructure-related aspects. In addition, over recent years one re-
search line has emerged that addresses the issue of “multilocality” of the people 
involved as well as their families (Weichhart 2009). Studies that concentrate on this 
topic are oftentimes geared towards one specifi c segment of circular mobility, such 
as weekend commuting or long-distance relationships (Collmer 2005; Schier 2009; 
Reuschke 2010). 

1 Every four years, commuting behaviour is studied in the microcensus (Mikrozensus). These 
studies concentrate on information such as distance between the residence and the workplace, 
the amount of time required for a commute and the most commonly used means of transporta-
tion (Grau 2009). In the surveys of income and consumption (Einkommens- und Verbrauchs-
stichprobe, EVS), information is gathered relating to second residences (Statistisches Bunde-
samt 2005). The Socio-economic Panel (Sozio-oekonomisches Panel, SOEP) also includes a 
segment on commuting. The survey looks into how often an employee commutes between his/
her main place of residence and the workplace as well as the distance between the workplace 
and the place of residence (Wagner 1989; Stutzer/Frey 2008). In their panel study, Huinink and 
Kley (Kley 2009) look into not only migration behaviour in general, but also into commuting mo-
bility in the cities of Freiburg and Magdeburg; in their commuter report, Wiethölter et al. (2009) 
focus on the Berlin-Brandenburg region. 



•    Heiko Rüger, Michael Feldhaus, Katharina S. Becker, Monika Schlegel224

The “Panel Analysis of Intimate Relationships and Family Dynamics” (pairfam) 
is another representative study, which resembles the study by Schneider and his 
colleagues with respect to its methods of recording circular mobility, and which 
focuses on the impact it has on developments relevant to partnerships and family. 
Provided in the section below is a brief introduction to the two studies.

Job Mobilities and Family Lives in Europe (JobMob)2

In a representative survey conducted in Germany in 2007, 1,495 individuals between 
the ages of 25 and 54 were randomly selected and interviewed about their experi-
ence with job mobility, with the help of a standard survey instrument via landline te-
lephone (CATI).3 In a second step, a random screening procedure was implemented 
to survey another 168 professionally mobile individuals (oversampling) in order to 
allow for differentiated analyses based on a wider empirical basis.4 Thus, a total of 
1,663 respondents were interviewed in the German random survey (for more details 
on this, see Schneider et al. 2008; Huynen et al. 2008, 2010).5

In terms of its content, the study focused on the following three main aspects (Lim-
mer/Schneider 2008): 

(1) What are the prevalence and variety patterns of spatial job mobility in Eu-
rope?

(2) What are the causes and circumstances of their emergence? 

(3) What are the consequences of job mobility for one’s partnership, family, sub-
jective well-being, career, social relationships and social networks? 

2 The research project “Job Mobilities and Family Lives in Europe” was fi nanced by the Europe-
an Commission from funds made available through the Sixth Framework Programme, priority 
“Citizens and governance in a knowledge-based society.”

3 In addition to the survey conducted in Germany, the survey was also conducted in fi ve other 
European countries (Spain, France, Belgium, Switzerland and Poland). A total of 7,220 people 
were interviewed in this survey; of these 2,432 were mobile.

4 Randomly selected individuals were surveyed in short interviews (screenings) about their mo-
bility behaviour and if these persons met a set of predefi ned criteria, they were also included in 
the sample (oversampling). 

5 The data were subjected to a design and an adjustment weighting. In a fi rst step, a design weight 
was calculated, which a) adjusts the imbalance of the probability of individuals from small and 
large households to be included in the sampling and b) adjusts the disproportionate number 
of professionally mobile persons interviewed in the representative part of the sampling. Based 
on the results, the bias that was calculated in the selective non-response was corrected. Five 
criteria were taken into account: age, gender, education, marital status and region (East/West). 
Offi cial statistics from 2007 were used as a reference. Number of cases prior to weighting: N 
(total) =1,663 and n (employed persons) =1,359. A detailed description can be found in Huynen 
et al. (2008).



Circular job-related spatial mobility in Germany    • 225

In 2010, a second interview phase was conducted in Germany. As a result, ad-
ditional information will be made available for future longitudinal analyses.6

Panel Analysis of Intimate Relationships and Family Dynamics (pairfam)7

The “Panel Analysis of Intimate Relationships and Family Dynamics“ (pairfam) project 
is a representative, interdisciplinary longitudinal study that aims at researching dif-
ferent life styles of people living in families or relationships in the Federal Republic 
of Germany. This 14-year long-term project aims at interviewing a group of ran-
domly selected individuals once a year. At the time of the fi rst survey in 2008/2009, 
these individuals were categorised into representative groups between the ages of 
15 to 17, 25 to 27, and 35 to 37 (cohort sequential design; N=12,402).8 Every year, 
the partner of the representative respondent (anchor person), if applicable, and/or 
from the second wave on their parents or step-parents and one children living in the 
household are also interviewed (multi-actor-design). The intended purpose of the 
pairfam project is to fi nd answers to the following questions: 

(1) What course do relationships take from the moment they are entered into un-
til the couple potentially divorces or separates? What are the consequences 
for the respective partners and their children (if there are any)? 

(2) What are the determining factors when founding and expanding one’s fam-
ily? 

(3) What are inter-generational relationships and the parent-child relationships 
like?

(4) What infl uence do the surrounding social networks have on the dynamics of 
a family or relationship? 

In addition, the modulated survey design that was used offers a number of dif-
ferent options in analysing the recorded data (cf. Huinink/Feldhaus 2008; Huinink et 
al. 2011). In the present paper, however, these will not be discussed further. Despite 
the fact that the analyses of job mobility patterns are not the major point of focus 
in the set of tools used in the pairfam survey, it does record central modules of job 
mobility in order to grasp its opportunity structure that is relevant for people living 
in relationships and families. Its approach to recording the data is similar to that of 
the fi rst Schneider et al. (2002a,/b) study. As a panel study, pairfam thus makes it 

6 The scientifi c use fi le of the dataset from the interview phase 2007 (1st wave) is available from 
GESIS – Leibniz-Institut für Sozialwissenschaften (http://www.gesis.org) with the study number 
ZA5065. You can fi nd more information about the European comparative research project at 
http://www.jobmob-and-famlives.eu.

7 The “Panel Analysis of Intimate Relationships and Family Dynamics” (pairfam) was funded by 
the German Research Foundation (Deutsche Forschungsgemeinschaft, DFG) as a long-term 
project. 

8 Two weighting variables were formed in the pairfam study: a design weighting that controls for 
the different cohort sizes (dweight) and an additional factor (psweight), which – on the basis of 
the microcensus – adjusts for the factors age, gender, marital status, and whether or not there 
are any children. For more information, refer to Brüderl et al. (2010). 



•    Heiko Rüger, Michael Feldhaus, Katharina S. Becker, Monika Schlegel226

possible to acquire data and analyse job mobility, its prerequisites, and the potential 
consequences that result from it with the help of a prospective cohort study design. 
Due to the comprehensive interview programme and the limited amount of time 
available for the interviews, detailed information on the subject of mobility is only 
available in every other study wave. Pairfam will thus be able to provide a compre-
hensive set of longitudinal data that can be used in mobility and family research in 
the future.9

2.2 Operationalisation of circular job-related spatial mobility 

As we mentioned before, whereas the JobMob survey had specifi cally been de-
signed to acquire data about job-related mobility patterns, job mobility was neither 
the main point of focus of the theoretical design, nor the focus of the pairfam survey. 
Yet, since the two surveys are largely based on the earlier study by Schneider and 
colleagues (Schneider et al. 2002a/b), many forms of circular job mobility can be 
operationalised in a more or less identical way. 

The conceptualisation of job mobility and the underlying categorisations are 
aimed at two things: fi rstly, capturing the existing types of job mobility as exhaus-
tively as possible, and secondly, distinguishing between the various requirements 
that might arise for the opportunity structure of social relationships (cf. sect. 1). The 
following is an overview of the operationalisation categories that were used.

In this context, the criteria used in the JobMob study serve as the starting point. 
The term mobile for job-related reasons refers to individuals who, a) commute long 
distances, i.e. for each journey to or from work require at least one hour and com-
mute at least three times a week; or b) have had to spend at least 60 nights away 
from home for job reasons in the past year. This category also includes Weekend 
commuters or such persons who, for job-related reasons, are in a long-distance 
relationship with at least one hour travel distance each way between their places of 
residence, or c) people who have relocated to a location at least 50 km away from 
their former residence at some point during the last three years. Individuals who are 
mobile in more than just one of these ways are referred to as ”Multi-mobile“ (Lim-
mer/Schneider 2008). Since this article focuses on circular mobility, we will only dis-
cuss points a) and b). In this context, we have defi ned two distinct concepts of job 
mobility, which are not mutually exclusive: job-related commuting and job-related 
overnight stays outside of a person’s place of residence.

Job-related commuting 

Job-related commuting is the result of an individual’s workplace being outside of 
their own place of residence, requiring them to travel to their workplace. In this 
article, we differentiate between three types of job-related commuters: (1) Short-

9 The datasets from the pairfam study can be requested from their subscriber service at http://
www.pairfam.de as a scientifi c use fi le. The website also provides additional information about 
the project.



Circular job-related spatial mobility in Germany    • 227

distance commuters: their commute (one way) takes between 1 and 29 minutes; 
(2) Medium-distance commuters: their commute (one way) takes between 30 and 
59 minutes; and (3) Long-distance commuters: the time required for their commute 
(one way) is at least 60 minutes. 

Up until now, there has been no differentiation into Short- and Medium-distance 
commuters within the scope of the JobMob study. In that study, the persons con-
cerned were not actually considered to be ”mobile” since their commute took com-
paratively little time.10 In this article, we have defi ned these different categories in 
order to be able to distinguish more clearly between the different commuters on 
the basis of their commute, and to be able to better contrast individuals with long 
commutes (Long-distance commuters) with those who have shorter ways to work 
(Short- and Medium-distance commuters). In addition, the empirical question that 
to date still remains unanswered is whether there are any differences within the 
group of individuals with shorter commutes. In both studies it is possible to imple-
ment these categorisations of the different types of commuters. The amount of time 
needed for a one-way commute to work is recorded in minutes.11

Job-related overnight travel

We have created the category of the so-called “Overnighters” to refer to individuals 
who frequently need to stay overnight outside of their place of residence for work 
reasons. Any individual who has spent at least 60 nights away from their primary 
place of residence for work reasons over the past 12 months can be referred to as 
an Overnighter. Individuals who fall into this category may either stay at varying 
work locations (Varimobiles) or in a second place of residence (Shuttlers). In both 
studies, the study respondents were asked to provide this information. Hence, two 
additional forms of circular mobility have been identifi ed: (4) Varimobiles: Individu-
als who have spent at least 60 nights away from home for job-related reasons over 
the last 12 months (e.g. in hotels, boarding houses or company housing) and (5) 
Shuttlers: Individuals who have a secondary residence close to their work location 
(especially Weekend commuters).

In order to operationalise Shuttlers, we assume the existence of the following 
basic criteria: (a) the existence of housing (e.g. secondary place of residence) close 

10 Since there were no established guidelines pertaining to the criteria of commuting time for 
Short-distance commuters and for Medium-distance commuters, these were defi ned by the 
authors in this case. As approximation the timespan has been divided in half.

11 One criterion was also that the person commuted at least three days each week. In the JobMob 
study, the frequency of a person’s commute was only recorded for the group of Long-distance 
commuters, whereas in the pairfam study, each and every respondent was queried consistently 
for all forms of commuting, whether they commuted on a daily basis or at least several times 
a week. On this issue, the pairfam study differentiates even further between whether the com-
mute takes place from the fi rst or the second place of residence. We do not make this distinction 
in our analyses, however.



•    Heiko Rüger, Michael Feldhaus, Katharina S. Becker, Monika Schlegel228

to the workplace that is considered the starting point of the commute to work,12 and 
(b) the number of times an individual spends the night at a place other than their 
primary place of residence/their base,13 and (c) for most of the time, the individual 
concerned must have worked at the same location, which constitutes another dif-
ference to Varimobiles. For the latter, varying work locations are a defi ning char-
acteristic. This can happen in the context of ”mobile professions,” such as profes-
sional drivers, pilots, or installation mechanics. 

The combined mobility indicator in the JobMob and pairfam studies

The ultimate design of the mobility indicators follows a strict systematic hierarchy. 
Scenarios involving multi-mobility, that is, in which an individual deals with at least 
two forms of mobility (cf. e.g. Limmer/Schneider 2008), are not explicitly indicated 
as such. Instead, the individuals concerned were assigned to a category on the 
basis of the following list of priorities: 1. Shuttlers, 2. Varimobiles, 3. Long-distance 
commuters, 4. Medium-distance commuters, 5. Short-distance commuters. When-
ever two forms of mobility applied, we selected the one that was expected to have a 
stronger impact on the opportunity structure of a family or relationship, that is, the 
one that placed higher demands on a person’s mobility. For example, Shuttlers are 
expected to live up to higher mobility demands than, for example, Short-distance 
commuters. In addition, there is another category of non-mobile individuals. In-
cluded in this category are those individuals who are gainfully employed at their 
place of residence (e.g. farmers, teleworkers). The category of “Other commuters” 
includes individuals whose commuting habits are unsystematic or irregular, so that 
they cannot be categorised on the basis of the criteria we provided. Due to the spe-
cifi c questionnaire used in the project, this group was only explicitly identifi ed in the 
JobMob study. It will not be treated as a specifi c category in any of the multivariate 
statistical analyses so as to ensure that study results remain comparable.14

12 In JobMob, the criterion was that this was kept due to the large distance between the workplace 
and the main place of residence; in pairfam, the number of times a person commuted to the 
workplace was assessed as seen from the secondary place of residence, which needed to have 
taken place at least once a week or at irregular intervals.

13 In JobMob, at least 60 nights over the last 12 months; in pairfam, data stating that the person 
had spent the night in a secondary place of residence at least once or twice a week and did not 
live there with a partner or with his/her parents.

14 Moreover, there are yet additional forms of mobility that can also be examined in the two stud-
ies. In addition to residential mobility, which is not the subject matter of this article, but which 
was also surveyed in the second wave of pairfam (the acquisition of a relocation biography in 
retrospect will be implemented in that study in a third wave), JobMob allows to analyse job-
related long-distance relationships. The latter is not possible in pairfam. In that study, there can 
merely be a differentiation as to whether couples are living together or not, without providing 
any reasons as to why they may be living together or not.



Circular job-related spatial mobility in Germany    • 229

3 Results

3.1 Descriptive analyses: the prevalence of circular job mobility and its 
correlation to socio-demographic characteristics

In a fi rst step, we will present the prevalence of circular job mobility in Germany on 
the basis of the two studies (cf. tab. 1 and 2). In a second step, the composition of 
the different types of mobility will be examined based on key socio-economic cri-
teria (cf. tab. 3 and 4). We assume that despite the different populations and objec-
tives used, the essential trends will become apparent in both studies, and that it will 
be possible to paint a complete picture of the phenomenon of circular job mobility 
in Germany. Guided by these assumptions, we will focus on a comparison between 
the two studies. For the pairfam study, only the age groups of 25 to 27-year-olds 
and 35 to 37-year-olds were selected, because for the 15 to 17-year-olds, job mobil-
ity is not (yet) relevant to the same degree. In the JobMob study, all respondents 
(between the ages of 25 and 54) were taken into account. The analyses were each 
calculated using the relevant adjustment weightings.15

Illustrated in table 1 is the distribution of different types of mobility according 
to the two studies. In this regard, we differentiate between two groups: in order to 
provide a complete overview of the mobility patterns represented in the popula-
tion sample of each study, we list all the information on the sampling as a whole 
(based on the ”overall population“ in each age group). In order to demonstrate the 
prevalence of the different types of mobility among the various kinds of gainfully 
employed persons, we provide numbers for each of these different groups individu-
ally as well. 

Despite differences between the two studies as concerns the age composition 
(pairfam: from the ages of 25 to 27 and 35 to 37; JobMob: between the ages of 
25 to 54) there are, in part, strong correlations between the distribution patterns. 
First of all, the percentage of overall job mobility – defi ned here as short-distance, 
medium-distance and long-distance commuting, varimobility, and weekend com-
muting (Shuttlers) – is higher in the JobMob study (75 %) than the 65 % that can 
be found in the pairfam study. As a result of the specifi c age groups chosen for the 
pairfam study (ages 25 to 27 and 35 to 37), there are signifi cantly more individuals 
who still had not fi nished their education or who were on parental leave, which is 
why the percentage of not gainfully employed persons is so much higher in that 
study. In contrast, when looking at the group of gainfully employed persons, the 
difference in overall mobility is but a minor one. The same applies if one takes a 
closer look at only those types of mobility for which increased mobility demands 
can be assumed, such as Long-distance commuters, Varimobiles, and Shuttlers: of 
all the respondents, they make up 11.7 % (JobMob) and 11.9 % (pairfam), respec-
tively; among gainfully employed persons, those categories make up 14.0 % and 

15  In JobMob, the weight ”w_nation“ was used, in pairfam, unless defi ned otherwise, the calcula-
tion was based on a combination of ”dweight“ and ”psweight“ (cf. sect. 2.1).



•    Heiko Rüger, Michael Feldhaus, Katharina S. Becker, Monika Schlegel230

16.4 %, respectively. These fi ndings clearly show the relevance of circular mobility 
arrangements, which can thus certainly not be considered to be only a marginal 
phenomenon in society.

However, differences with regard to the percentage of Short-distance commut-
ers become apparent. In the JobMob study, Short-distance commuters total 46.9 % 
of all the respondents (as compared to 36.8 % in pairfam). In contrast, though, if 

Tab. 1: Circular job mobility data from the JobMob and pairfam studies

 JobMob1 Pairfam 

 

All respondents Gainfully 
employed 
persons2 

All respondents Gainfully 
employed 
persons 

 n % n % n % n % 

Short-distance 
commuters 701 46.9 701 57.3 2611 36.8 2611 51.0 

Medium-distance 
commuters 241 16.1 241 19.7 1155 16.3 1155 22.5 

Long-distance 
commuters 97 6.5 97 8.0 364 5.1 364 7.1 

Varimobiles 68 4.5 65 5.3 448 6.3 448 8.7 

Shuttlers 11 0.7 9 0.7 33 0.5 33 0.6 

Non-mobiles3 59 3.9 59 4.8 512 7.2 512 10.0 

Other commuters4 51 3.4 51 4.2 - - - - 

Not gainfully 
employed5 267 17.9 - - 1964 27.7 - - 

Total 1,495 100 1,223 100 7,087 100 5,123 100 

1 The number of cases for the forms of mobility “Varimobile” and “Shuttlers” may devi-
ate slightly within the group of all respondents and the working population, because 
they were queried in retrospect with regard to spending the night away from home for 
job-related reasons for the last 12 months. 

2 Gainfully employed persons are such persons who are either employed full-time, are 
self-employed, employed part-time or are occasionally employed, those who are in 
training or are taking part in an internship programme, or who perform any other activ-
ity in return for compensation.

3 Included in the category of “non-mobile” persons are also those who make their living 
from home (e.g. farmers, teleworkers). 

4 Included in the category “other commuters” are also those persons with commuting 
behaviour described as extremely unsystematic or irregular, so that they cannot be 
grouped based on the criteria used. Because of the relevant questionnaire, this group 
can only be identifi ed in the JobMob study.

5 Included in the group of “Not gainfully employed” are those persons currently on pa-
rental leave, unemployed, homemakers or those in early retirement.

Source: Job Mobilities and Family Lives 2007; pairfam (Wave 1, 2008/2009); own calcula-
tions; weighted data



Circular job-related spatial mobility in Germany    • 231

we concentrate on the group of gainfully employed persons, these differences are 
already signifi cantly lower. Any remaining differences can probably be attributed 
to the fact that individuals exhibiting typical criteria of Short-distance commuters, 
such as part-time employees and women, and also, though less often, persons who 
live in major cities (cf. tab. 3 and 4), were represented in higher numbers in the Job-
Mob study than they were in pairfam. 

In both studies, the group of Short-distance commuters is the largest one, fol-
lowed by Medium-distance commuters (16.1 % in JobMob and 16.3 % in pairfam 
as related to the total number of respondents). The numbers of Long-distance com-
muters in both studies were almost the same, although the group of Varimobiles 
was slightly larger in the pairfam study. This is particularly true when referring only 
to the group of gainfully employed persons. However, overall, the picture was fairly 
homogenous as concerns the prevalence of individual types of circular mobility.

To allow for a direct comparison, the population sample in the JobMob study 
was adjusted to correspond to the age groups (ages 25-27 and 35-37) used in the 
pairfam study (cf. tab. 2). As a result of the differences in the study design, the num-
bers of respondents in both studies differ signifi cantly. For this specifi c age group, 
there are signifi cantly fewer cases available for analysis in the JobMob study. Ex-
cept for the age group of 25 to 27-year-olds in the JobMob study, the results from 
both studies are fairly similar. Moreover, the differences between the age groups 
in the pairfam study are minor. This result is surprising at fi rst, because different 
mobility patterns were in fact expected for each age group. For example, we know 
from earlier studies that mobility decreases with age, especially when taking into 

Tab. 2: Circular mobility in JobMob and in pairfam among gainfully employed 
persons, by age groups

 
 JobMob Pairfam1 

 Age 25-27 Age 35-37 Age 25-27 Age 35-37  

 n % n % n % n % 

Short-distance 
commuters 41 55.1 74 56.1 1,376 52.9 1,584 49.6 

Medium-distance 
commuters 11 15.1 22 22.0 584 22.5 722 22.6 

Long-distance 
commuters 12 16.3 8 5.8 182 7.0 229 7.2 
Varimobiles 7 9.5 7 5.0 223 8.6 284 8.9 

Shuttlers 0 0 2 1.7 20 0.8 17 0.5 

1 Since the groups are listed separately here, instead of the design weighting, only selec-
tion weighting (psweight) is used. This is why the (weighted) number of cases differs 
from what is shown in table 1.

Source: Job Mobilities and Family Lives 2007; pairfam (Wave 1, 2008/2009); own calcula-
tions; weighted data



•    Heiko Rüger, Michael Feldhaus, Katharina S. Becker, Monika Schlegel232

account residential types of mobility (e.g. Schneider et al. 2008; Wagner 1989). Con-
sequently, the minor differences between the age groups in the pairfam study do 
indicate fewer age differences in circular mobility patterns. 

In the following, we will conduct a comparative analysis of the correlation be-
tween mobility patterns and the different socio-demographic, regional, and part-
nership- as well as family-related criteria in both studies with the help of common 
consolidated indicators (cf. tab. 3 and 4).16 Considering the low number of Shuttlers, 
these have been included in the group of Varimobiles. Both of these groups com-
bined constitute the category of Overnighters (cf. sect. 2.2).17

From previous studies, which were mostly based on the data found in the Job-
Mob study, we know that the various socio-demographic groups do, in part, ex-
hibit signifi cant differences in mobility behaviour. For example, gainfully employed 
women are signifi cantly less likely to travel as Long-distance commuters or Over-
nighters than is the case with men, presumably because more women dedicate 
themselves to housework and parenting than men do (e.g. Schneider et al 2002a/b; 
Meil 2010a/b). Furthermore, there are also indications that some forms of job mobil-
ity do correlate with an individual’s education, the form of employment (part-time 
or full-time), and the individual’s occupational prestige (e.g. Schneider et al. 2008; 
Ruppenthal/Lück 2009; Lück/Ruppenthal 2010). 

By using comparative indicators, we will analyse in the following whether these 
correlations can be reproduced in a similar fashion in the pairfam study. Moreover, 
we will analyse information relating to the geographic context. Due to differences 
relating to the labour market and infrastructure, we should be able to fi nd differ-
ences in the distribution structures between the different groups in eastern and 
western Germany (higher mobility demands apply in eastern Germany) as well as 
between groups in major cities and in rural areas/small towns. We will fi nally exam-
ine whether mobility arrangements differ, depending on partnership- and family-
related criteria. For example, the assumption could be made that persons who need 
to deal with higher mobility demands are more limited in their ability to fi nd a part-
ner (Rüger/Ruppenthal 2010). At the same time, however, the fact that a person is in 
a relationship is often the reason for a specifi c mobility arrangement, for example, 
as a way to reconcile two professional careers (Rüger/Becker 2011). Therefore, the 
question that arises is whether there are differences between the different types of 
mobility in terms of a person’s relationship status. Moreover, one could also argue 
that persons who give in to higher mobility demands might be less strongly tied to 
relationship or family (Sennett 1998), and therefore less likely to be in a commit-
ted relationship, i.e. be married. Relationships that are subjected to higher mobility 
demands may be more selective, meaning that individuals who have to deal with 
higher mobility demands show a tendency to have shorter relationships. Conse-

16 The category of „Other commuters“ is not taken into account because there is no way to com-
pare the two studies in this analysis (cf. sect. 2.2).

17 Because the group of „Overnighters“ includes signifi cantly more varimobile persons than Shut-
tlers, the relevant fi ndings are more strongly considered for the fi rst group. 



Circular job-related spatial mobility in Germany    • 233

quently, they may not yet have contemplated the possibility of entering into a more 
institutionalised relationship (such as marriage). 

By the same token, a person may fi rst want to meet certain basic conditions be-
fore deepening their relationship with another, such as having more time to spend 
together for planning their future. Self-imposed prerequisites of this kind, however, 
might be thwarted because of the couple’s current mobility patterns. Similar argu-
ments might apply to family development. As far as that is concerned, the JobMob 
data shows a correlation between an individual’s mobility patterns and childless-
ness (e.g. Meil 2010a/b; Rüger/Becker 2011). We will fi rst conduct a bivariate analy-
sis to fi nd an answer to the question of whether these and other tendencies could 
also be confi rmed in the pairfam study results. In the next section, these assump-
tions will be examined based on criteria relevant to partnerships and family by ap-
plying multivariate analyses.

An assessment based on the age criteria is not directly comparable in the two 
studies. However, in both studies the relationship between the different mobility 
arrangements and the age groups (V=0.084 and V=0.049) is minimal. Based on the 
JobMob study, a trend can be identifi ed (cf. tab. 3), whereby job-related commuting 
with longer commute times (long-distance commuting) is disproportionately high 
both among younger individuals (ages 25-34) and in the group of older respondents 
(ages 45-54). However, according to the JobMob study, middle-aged respondents 
(ages 35-44) are overrepresented in the category of Overnighters (Varimobiles and 
Shuttlers). As concerns the correlation between mobility and gender, both studies 
offer very similar results. The longer the commuting time, the higher the proportion 
of men in the respective group of people, or rather the more likely it is for the pro-
portion of women to decrease. These results match the fi ndings shown in the mi-
crocensus of 2004 (Statistisches Bundesamt 2005). In both studies, Long-distance 
commuters and Overnighters are clearly more often men. Thus, the general fi nding 
according to which job mobility is usually a ”male phenomenon“ (e.g. Schneider et 
al. 2008) could be confi rmed. 

These results were somewhat more pronounced in the JobMob study than the 
pairfam study. One explanation for the specifi c gender differences can be found 
in the different lifestyles and working conditions of men and women. Despite the 
growing signifi cance of women in the labour market, women in Germany are still 
more often the partner responsible for housework and childcare (e.g. Peuckert 2008). 
Part-time positions, which often make it easier to balance family life and career, and 
which require job mobility far less often than full-time positions do (e.g. Rüger/Beck-
er 2011), are consequently more often held by women than by men. Furthermore, a 
disproportionately higher number of men work in ”mobile professions”.

Regarding formal educational status, both studies show some similar correla-
tions. In both studies, individuals with an elementary school or secondary school 
education are represented slightly more often among the group of Short-distance 
commuters. In addition, individuals with higher levels of education can mainly be 
found among those with a medium-distance commute. Particularly individuals with 
a university degree are more likely to be found in those groups with higher mobil-
ity demands compared to the group of Short-distance commuters. All in all, due to 



•    Heiko Rüger, Michael Feldhaus, Katharina S. Becker, Monika Schlegel234

Tab. 3: Forms of circular mobility in the JobMob study, composition according 
to key socio-demographic characteristics (in %)

 Total Non-
mobiles 

Short-
distance 

commuters 

Medium-
distance 
comm. 

Long-
distance 

commuters 

Over-
nighters 

V 
(p value)1 

Age        

25-34 24.1 14.3 23.6 25.3 30.5 23.3 

35-44 40.6 37.5 41.8 37.8 30.5 53.4 

45-54 35.3 48.2 34.5 36.9 38.9 23.3 

0.084 
(0.035) 

Gender        

Male 55.8 57.6 50.9 57.7 63.9 83.8 

Female 44.2 42.4 49.1 42.3 36.1 16.2 

0.169 
(<0.001) 

Education        

Elementary school 22.1 20.7 23.8 14.7 20.9 18.9 

Secondary school 47.0 44.8 48.7 43.5 51.6 37.8 

Upper secondary school 13.5 12.1 12.9 16.8 9.9 14.9 

University 18.3 22.4 14.5 25.0 17.6 28.4 

0.092 
(0.005) 

Employment status        

Full-time 78.7 79.2 75.5 79.2 87.4 95.7 

Part-time 21.3 20.8 24.5 20.8 12.6 4.3 

0.134 
(<0.001) 

Occupational prestige        

ISEI (16-40) 43.6 28.1 49.0 32.2 39.8 46.2 

ISEI (41-65) 38.5 45.6 33.3 50.6 46.2 32.3 

ISEI (66-90) 17.9 26.3 17.6 17.2 14.0 21.5 

0.125 
(<0.001) 

Geographic factors        

West 85.8 89.7 89.0 77.9 83.0 82.2 

East 14.2 10.3 11.0 22.1 17.0 17.8 

0.131 
(0.001) 

Population 100,000 or more 34.5 28.1 29.4 51.3 30.7 37.5 

Population 20,-100,000 26.4 42.1 29.0 18.9 20.4 19.4 

Population up to 20,000 39.2 29.8 41.5 29.7 48.8 43.1 

0.148 
(<0.001) 

Relationship- and family-related factors      

Partner 79.6 74.6 79.7 79.7 81.4 80.6 0.032 
       (0.883) 

Married 54.1 64.4 55.2 51.0 52.1 48.6 0.064 
       (0.307) 

Childless 32.8 17.2 31.8 35.7 38.1 38.9 0.093 
       (0.040) 

1 Calculation based on Chi-square and Cramer’s V tests.

Source: Job Mobilities and Family Lives 2007; own calculations; weighted data; weighted 
N≥1,122



Circular job-related spatial mobility in Germany    • 235

Tab. 4: Forms of circular mobility in the pairfam study, composition according 
to key socio-demographic characteristics (in %) 

Source: pairfam (Wave 1, 2008/2009); own calculations; weighted data; weighted 
N=5,123, (the analyses refer to both age groups, which is why the combined 
weight dxpsweight was used here, see Brüderl et al. 2010).

 Total Non-
mobiles 

Short-
distance 

commuters 

Medium-
distance 
comm. 

Long-
distance 

commuters 

Over-
nighters 

V 
(p value) 

Age        

25-27 42.0 34.6 43.6 41.8 41.5 41.9 

35-37 58.0 65.4 56.4 58.2 58.5 58.1 

0.053 
(0.006) 

Gender        

Male 55.6 55.8 50.7 56.6 60.2 76.3 

Female 44.4 44.2 49.3 43.4 39.8 23.7 

0.148 
(<0.001) 

Education        

Elementary school 18.4 19,8 20,7 14,5 17,9 14,5 

Secondary school 36,5 32,6 38,9 33,7 33,8 36,7 

Upper secondary degree 19,6 22,0 18,3 21,6 19,6 19,1 

University 25,4 25,5 22,1 30,2 28,8 29,8 

0.065 
(<0.001) 

Employment status        

Full-time 79.6 74.1 76.4 80.5 86.5 95.6 

Part-time 20.3 25.9 23.6 19.5 13.2 4.4 

0.169 
(<0.001) 

Occupational prestige        

ISEI (16-40) 46.8 49.5 49.3 41.8 37.7 48.7 

ISEI (41-65) 36.7 40.7 36.6 36.5 41.6 29.7 

ISEI (66-90) 16.6 9.8 14.1 21.7 20.8 21.5 

0.126 
(<0.001) 

Geographic factors        

West 83.2 85.5 85.2 82.3 78.8 75.5 

East 16.8 14.5 14.8 17.7 21.2 24.5 

0.084 
(<0.001) 

Population 100,000 or more 67.1 70.3 65.2 72.0 70.5 59.1 

Population 20,-100,000 20.9 18.9 22.3 17.4 16.0 27.3 

Population up to 20,000 12.0 10.7 12.5 10.6 13.5 13.6 

0.062 
(<0.001) 

Relationship- and family-related factors     

Partner 76.2 73.4 77.6 74.0 75.6 76.9 0.041 
       (0.077) 

Married 53.8 58.1 54.6 53.2 50.2 48.8 0.047 
       (0.073) 

Childless 57.4 51.4 55.4 62.5 61.2 60.1 0.073 
       (<0.001) 



•    Heiko Rüger, Michael Feldhaus, Katharina S. Becker, Monika Schlegel236

the modestly strong correlations among the two samplings, it seems that groups 
organised by educational background obviously exhibit fewer differences in terms 
of their circular mobility behaviour than with regard to their residential mobility pat-
terns (see e.g. Schneider/Meil 2008; Schneider/Collet 2010).

As concerns the employment status (part-time or full-time), both studies show 
very similar fi ndings. The higher the mobility demands, the lower the proportion of 
persons working in part-time employment. Whereas especially among the group 
of Short-distance commuters, of which close to a quarter are part-time employees, 
this number is still slightly disproportionate, the percentage decreases all the way 
down to 4 % in the group of Overnighters. There are also some very clear patterns 
concerning occupational prestige. Since there is a positive correlation between 
commuting distances and level of income (Statistisches Bundesamt 2005), the 
number of Short-distance commuters was expected to be higher among respond-
ents with a very low professional status as compared to the Medium-distance and 
Long-distance commuters. However, it is remarkable that both studies show a slight 
overrepresentation of individuals with the lowest but also with a higher professional 
standing in the group of Overnighters. Included here are occupations such as lorry 
drivers and assembly workers, who, despite their lower occupational status, must 
accept the fact that they are Overnighters by job. This observation points to a cer-
tain amount of heterogeneity within this form of mobility; this assumption had al-
ready been made by other studies (e.g. Schneider et al. 2009b; Nolle 2005). 

When looking at regional differences between eastern and western Germany, 
it seems obvious that the hypothesis expressed above can be confi rmed by both 
studies: While mobility requirements are less pronounced in the west of Germany, 
in the east Long-distance commuters and Overnighters are more common. This 
might result from different infrastructural systems as well as settlement patterns. In 
some cases, however, an increased ”pressure to be mobile“ on the labour market is 
the reason. This was further confi rmed by previous fi ndings, according to which (in 
eastern Germany especially), there are very few people who rejected the mobility 
requirement (Schneider et al. 2008). With regard to the correlation between certain 
mobility types and the number of inhabitants of a town or city, there are signifi cant 
differences in marginal distribution: in the pairfam study, the largest group (67 %) 
lives in major cities with a population of 100,000 and more (in contrast to only 35 % 
in the JobMob study). In the JobMob study, the largest group of respondents lives 
in locations with population sizes of up to 20,000; this group in turn only made up 
12 % in the pairfam study. Despite these differences in marginal totals which are 
caused by different populations, there are in part similar tendencies within the dif-
ferent regions by population size. In major cities, those people with shorter com-
muting times are underrepresented because of the characteristic inner-city traffi c 
conditions, whereas the percentage of those with medium-long commuting times 
is higher as compared to the overall distribution. Even though in both studies re-
spondents from smaller cities (20,000-100,000 people) were represented less of-
ten in the group of Long-distance commuters, the JobMob study showed that a 
comparatively high share of respondents from small towns or rural areas are Long-
distance commuters. The majority of these individuals are most likely persons ages 



Circular job-related spatial mobility in Germany    • 237

40 and up who live ”in the country“ with their children and are willing to accept long 
commutes. However, the number of individuals who spend the night away from 
home varies between the studies; in the pairfam study, the majority of these live in 
medium-sized cities. In the JobMob study, results were exactly the opposite. One 
explanation for these fi ndings could be that different, age-specifi c professions are 
represented in the group of Overnighters, which rather correlate with geographic/
regional factors.  

When looking at partnership- and family-related characteristics, it seems that 
the share of individuals in relationships is about the same in both studies. People’s 
partnership status seems to have only a minor infl uence on the type of mobility 
they represent. As far as an individual’s marital status is concerned, there were 
only minor differences between mobility types in the two studies, as well; the only 
conclusion that can be derived, if any, is that in comparison to the overall popula-
tion, persons with higher mobility demands, such as Long-distance commuters and 
Overnighters, are less likely to be married. Differences in childlessness were much 
more pronounced: there was a positive correlation between long commutes/high 
mobility requirements and childlessness. As can be expected, there are large dif-
ferences between the studies in the number of people that were childless. In this 
regard, the pairfam study exhibits signifi cantly higher numbers of childlessness due 
to its focus on younger age groups. 

The potential relevance of circular job mobility for partnership- and family-relat-
ed aspects will be examined more closely in the following section by conducting 
multivariate analyses.

3.2 Multivariate analyses: the ability to explain aspects of partnership 
and family development based on circular job-related spatial 
mobility

In addition to the descriptive analyses, which are merely able to supply an initial 
point of reference, comparative multivariate analyses will be conducted. From these 
analyses we expect to obtain differentiated results on a number of different issues. 
The objective is to examine whether different mobility arrangements have an ef-
fect on an individual’s relationship status, the institutionalisation of a relationship, 
and family development in general. In particular, we will examine the correlation 
between different types of mobility and the likelihood of being in a committed re-
lationship, being married, and having at least one child in the current relationship. 
The focus here is not to fi nd any extensive explanation for an individual’s relation-
ship or family status. Rather, we want to fi nd out by means of comparative analyses 
whether an individual’s mobility patterns are correlated with the stated variables. 
Can the fi ndings of the descriptive analyses be upheld if we control for key covari-
ates that are linked to mobility patterns? It is absolutely possible, for instance, that 
the descriptive results shown may in fact be the result of correlations with other 
criteria rather than with the mobility status. Thus, in order to make our models more 
specifi c, it is crucial to include such covariates that are linked to circular mobil-
ity behaviour. That way, we will be able to gauge the effects of the various forms 



•    Heiko Rüger, Michael Feldhaus, Katharina S. Becker, Monika Schlegel238

of mobility as precisely as possible. Another objective is to fi nd a simple way to 
structure the models so that they can be properly compared. Accordingly, we must 
incorporate those socio-demographic as well as spatial-structural factors that have 
already been shown as relevant in our descriptive analyses (cf. sect. 3.1). Moreover, 
we will model the interaction between mobility patterns and gender, in order to be 
able to illustrate any gender-related differences. Studies conducted in the past have 
repeatedly shown that mobility may involve different conditions and consequences 
for men and women, respectively (e.g. Schneider et al. 2002a/b; Meil 2010a/b; Feld-
haus/Schlegel 2011; Rüger/Becker 2011). This aspect will be examined with regard 
to the target variables under study in this paper. 

The basic sample population for all three issues under review constitute gain-
fully employed persons. When analysing the relationship status of participants, all 
respondents will be included in the survey (N=1,297 in the JobMob study; N=5,686 
in the pairfam study). As far as the JobMob study is concerned, the category of 
”Other commuters“ will be excluded from the analyses, so that it can be compared 
to the pairfam study (cf. sect. 2.2). For reasons of homogeneity, we will only take 
into account individuals who are currently cohabiting with their partner (N=971 or 
841 for JobMob; N=4,390 for pairfam) when analysing the respondents’ family de-
velopment (childless or not) and the degree to which partnerships are institution-
alised (marital status). Since their number is small, we will include the Shuttlers in 
the Overnighters group together with varimobile individuals (cf. sect. 2.2). Target 
variables are dichotomous details about whether the participant in question is in 
a relationship, whether the respondent is married, or whether the individual has 
children with their current partner (codes: no=0/1=yes).18 We will calculate binary-
logistic regression models. The reference category relating to the other mobility 
forms is made up of the group of Short-distance commuters in each case. The odds 
ratios (eb) are shown in the tables.19

Relationship status

With regard to the relationship status, the multivariate analyses do initially confi rm 
the descriptive fi ndings. Neither in the JobMob nor in the pairfam study has it be-
come apparent that different types of mobility have any signifi cant effect on the 
likelihood of having a partner (cf. tab. 5 and 6, model 1). In contrast, however, when 
including the interaction term into the analyses (cf. tab. 5 and 6, model 2), some 
signifi cant effects become apparent, albeit weak ones. When comparing women 

18 For pairfam, the attribution of whether children exist is provided as an exact variable, in Job-
Mob, any children born after the current relationship began are attributed to this relationship.

19 As to the interpretation: A value above 1 shows a positive infl uence of the variables incorpo-
rated, a value below 1 indicates a negative infl uence (for categorical characteristics each as 
compared to the selected reference category). A value of 1.5 means, for example, that if an 
independent variable in the measuring unit the comparison is based on increases by one unit, 
the relative odds that the focused events might in fact occur is increased by a multiplier of 1.5 
(Backhaus et al. 2003; Jaccard 2001).



Circular job-related spatial mobility in Germany    • 239

who are Medium-distance commuters, Long-distance commuters or Varimobiles to 
women who are Short-distance commuters, the odds of them living in a commit-
ted relationship are lower as compared to the same comparison amongst men; this 
can be understood to mean that women with higher mobility arrangements are less 
likely to live with a partner as compared to women with lower mobility demands, or 
men. Furthermore, in the JobMob study, the observation has been made that men 
who commute long distances are more likely to live with a partner than men who 
commute short distances. Considering the coeffi cient for the gender effect (while at 
the same time accounting for the interaction term based on gender times mobility), 
it expresses the ratio between women and men (provided that the mobility vari-
able equals zero, i.e. in this case both are Short-distance commuters). The pairfam 
study illustrates a positive correlation between women commuting short distances 
and the likelihood of them having a partner, whereas this correlation is not as pro-
nounced for male Short-distance commuters (tab. 6, models 2 and 3). 

Marital status

When looking at an individual’s marital status and its correlation to their mobility 
behaviour, the two studies show similar results. Model 4 (tab. 5 and 6) illustrates 
a positive correlation between non-mobile individuals and their marital status, as 
compared to Short-distance commuters. In addition, in both studies there is a sig-
nifi cant negative correlation between Overnighters and married persons; more-
over, the correlations between any other types of mobility and the marital status, 
as compared to the reference category of short-distance commuting, are negative. 
If the interaction effect is included (cf. tab. 5 and 6, model 5), what is apparent, es-
pecially with regard to the pairfam data, is the gender-specifi c correlation that was 
assumed. Whereas the mobility patterns among men do not have an impact on the 
likelihood of whether or not they are married, women with more demanding mobil-
ity arrangements are married less frequently, both when comparing them to men in 
the same constellation and when comparing them to women who commute short 
distances. If we check further for any additional factors of relevance, at least the 
effect for women within the group of non-mobile persons as well as in the group of 
Overnighters remains unchanged (cf. tab. 6, model 6). In the JobMob study, the ten-
dencies and the degree to which these tendencies manifest themselves are almost 
the same as those above. Due to the lower number of cases, however, they are not 
statistically signifi cant.

Childlessness in current relationships

For the criterion of being childless (in the current relationship), the multivariate anal-
yses confi rm the descriptive results, even when checking for additional key factors 
of infl uence. In both studies, we fi nd that effects are greater for more extensive mo-
bility arrangements than for the reference group of Short-distance commuters. In 
contrast, non-mobile individuals are signifi cantly less likely to be without children, 
as could be expected. When also accounting for the interaction term (cf. tab. 5 and 



•    Heiko Rüger, Michael Feldhaus, Katharina S. Becker, Monika Schlegel240

T
a

b
. 

5:
 

C
ir

cu
la

r 
jo

b
 m

o
b

ili
ty

 a
n

d
 t

h
e
 c

o
rr

e
la

ti
o

n
 t

o
 a

n
 in

d
iv

id
u

a
l’s

 r
e

la
ti

o
n

sh
ip

 s
ta

tu
s,

 m
a
ri

ta
l s

ta
tu

s 
(w

h
e

n
 li

v
in

g
 w

it
h

 
a 

p
a
rt

n
e

r)
, 
an

d
 c

h
ild

le
ss

n
e

ss
 (

w
h

e
n

 li
v

in
g

 w
it

h
 a

 p
a
rt

n
e

r)
 a

m
o

n
g

 g
a
in

fu
lly

 e
m

p
lo

y
e

d
 p

e
rs

o
n

s 
(o

d
d

s 
ra

ti
o

s)
 –

 
Jo

b
M

o
b

 d
a
ta

S
o

u
rc

e
: 

Jo
b

 M
o

b
ili

ti
e
s 

an
d

 F
am

ily
 L

iv
e
s 

2
0
0
7
; 

u
n

w
e
ig

h
te

d
 d

at
a;

 o
w

n
 c

al
cu

la
ti

o
n

s;
 b

in
ar

y
 lo

g
is

ti
c 

re
g

re
ss

io
n

s;
 +

=
p

<
0
.1

0
; 

*=
p

<
0
.0

5
; 

**
=

p
<

0
.0

1
; 

**
*=

p
<

0
.0

0
1
; 

(a
) 

C
o

-e
ffi

 c
ie

n
t 

ca
n

n
o

t 
b

e
 d

e
te

rm
in

e
d

 b
e
ca

u
se

 t
h

e
 t

ar
g

e
t 

v
al

u
e
 i
s 

a 
co

n
st

an
t 

(i
.e

. 
al

l 
n

o
n

-m
o

b
ile

 
w

o
m

e
n

 a
re

 m
ar

ri
e
d

 o
r 

ar
e
 m

o
th

e
rs

)

 
R

e
la

ti
o

n
sh

ip
 (

0
/1

) 
M

ar
ri

e
d

 (
0
/1

) 
C

h
ild

le
ss

 (
0
/1

) 
 

M
o

d
e
l 1

 
M

o
d

e
l 2

 
M

o
d

e
l 3

 
M

o
d

e
l 4

 
M

o
d

e
l 5

 
M

o
d

e
l 6

 
M

o
d

e
l 7

 
M

o
d

e
l 8

 
M

o
d

e
l 9

 

N
o

n
-m

o
b

ile
 (

R
e
f.

: 
S

h
o

rt
 -

d
is

ta
n

ce
 c

o
m

m
u

te
rs

) 
.7

8
1
 

.5
2
2
+

 
.5

3
4
 

2
.3

6
8
* 

.8
0
4
 

.6
5
2
 

.2
7
8
**

 
.5

1
3
 

.5
8
8
 

M
e
d

iu
m

-d
is

ta
n

ce
 c

o
m

m
. 
(R

e
f.

: 
S

h
o

rt
-d

is
ta

n
ce

 
co

m
m

u
te

rs
) 

.9
5
3
 

1
.0

3
0
 

.9
9
5
 

.8
9
9
 

1
.0

3
8
 

1
.3

4
1
 

1
.2

1
1
 

.8
8
1
 

.7
4
6
 

Lo
n

g
-d

is
ta

n
ce

 c
o

m
m

. 
(R

e
f.

: 
S

h
o

rt
-d

is
ta

n
ce

 
co

m
m

u
te

rs
 )

 
1
.1

9
7
 

1
.9

0
1
+

 
1
.9

4
1
+

 
.7

2
4
 

1
.0

1
5
 

1
.0

3
5
 

1
.7

4
5
* 

.9
7
7
 

1
.0

0
8
 

O
v
e
rn

ig
h

te
rs

 (
R

e
f.

: 
S

h
o

rt
-d

is
ta

n
ce

 c
o

m
m

u
te

rs
) 

1
.0

1
7
 

1
.4

2
2
 

1
.3

0
1
 

.5
5
7
* 

.6
8
3
 

.9
0
7
 

1
.5

9
0
+

 
.9

8
0
 

.7
8
3
 

N
o

n
-m

o
b

ile
*w

o
m

e
n

 
 

2
.6

2
6
 

2
.1

7
0
 

 
(a

) 
(a

) 
 

.1
9
3
 

(a
) 

M
e
d

iu
m

-d
is

ta
n

ce
 c

o
m

m
.*

w
o

m
e
n

 
 

.8
7
9
 

.9
5
0
 

 
.7

7
6
 

.7
2
7
 

 
1
.7

4
7
 

1
.6

7
2
 

Lo
n

g
-d

is
ta

n
ce

 c
o

m
m

.*
w

o
m

e
n

 
 

.4
4
6
+

 
.4

7
8
 

 
.5

1
1
 

.6
5
1
 

 
3
.1

4
5
* 

3
.3

4
9
+

 
O

ve
rn

ig
h

te
r*

w
o

m
e
n

 
 

.3
7
9
+

 
.5

3
5
 

 
.4

6
3
 

.5
8
7
 

 
4
.4

4
0
* 

3
.9

8
0
+

 

W
o

m
e
n

 (
R

e
f.

: 
M

e
n

) 
.9

1
2
 

1
.0

3
2
 

.8
9
6
 

.9
6
8
 

1
.0

7
6
 

.7
8
7
 

1
.0

7
1
 

.7
6
6
 

1
.5

0
2
 

A
g

e
 3

5
-4

4
 (

R
e
f.

: 
2
5
-3

4
) 

 
 

1
.6

2
6
**

 
 

 
5
.5

1
1
**

*
 

 
.0

8
5
**

*
A

g
e
 4

5
-5

4
 (

R
e
f.

: 
2
5
-3

4
) 

 
 

1
.6

3
6
**

 
 

 
9
.2

5
7
**

*
 

 
.0

4
9
**

*
C

o
m

p
le

te
d

 s
e
co

n
d

ar
y
 s

ch
o

o
l (

R
e
f.

: 
e
le

m
e
n

ta
ry

) 
 

 
.6

7
0
 

 
 

.8
0
7
 

 
 

1
.7

6
5
 

U
p

p
e
r 

se
co

n
d

ar
y
 s

ch
o

o
l (

R
e
f.

: 
e
le

m
e
n

ta
ry

) 
 

 
.9

0
5
 

 
 

.5
5
1
+

 
 

 
3
.3

3
6
**

U
n

iv
e
rs

it
y
 d

e
g

re
e
 (

R
e
f.

: 
e
le

m
e
n

ta
ry

) 
 

 
1
.0

3
4
 

 
 

.6
8
4
 

 
 

3
.7

9
9
**

IS
E

I 
(4

1
-6

5
) 

(R
e
f.

: 
IS

E
I 
1
6
-4

0
) 

 
 

1
.0

1
1
 

 
 

1
.1

2
2
 

 
 

.7
9
2
 

IS
E

I 
(6

6
-9

0
) 

(R
e
f.

: 
IS

E
I 
1
6
-4

0
) 

 
 

1
.0

6
8
 

 
 

.9
9
5
 

 
 

.7
9
9
 

W
e
st

 (
R

e
f.

: 
E

a
st

) 
 

 
1
.1

0
7
 

 
 

1
.1

6
1
 

 
 

5
.2

2
3
**

*
F
u

ll-
ti

m
e
 (

R
e
f.

: 
p

a
rt

-t
im

e
) 

 
 

2
.1

8
5
**

*
 

 
2
.7

7
5
**

*
 

 
.1

4
6
**

*

N
a
g

e
lk

e
rk

e
 R

2
 

0
.0

0
3
 

0
.0

1
4
 

0
.0

5
7
 

0
.0

2
0
 

0
.0

4
7
 

0
.2

8
7
 

0
.0

3
1
 

0
.0

5
5
 

0
.4

8
0
 



Circular job-related spatial mobility in Germany    • 241

T
a

b
. 

6:
 

C
ir

cu
la

r 
jo

b
 m

o
b

ili
ty

 a
n

d
 t

h
e
 c

o
rr

e
la

ti
o

n
 t

o
 r

e
la

ti
o

n
sh

ip
 s

ta
tu

s,
 m

a
ri

ta
l s

ta
tu

s 
(w

h
e

n
 li

v
in

g
 w

it
h

 a
 p

a
rt

n
e

r)
 a

n
d

 
ch

ild
le

ss
n

e
ss

 (
w

h
e

n
 li

v
in

g
 w

it
h

 a
 p

a
rt

n
e

r)
 a

m
o

n
g

 g
a
in

fu
lly

 e
m

p
lo

y
e

d
 p

e
rs

o
n

s 
(o

d
d

s 
ra

ti
o

s)
 –

 p
a
ir

fa
m

 d
a
ta

 (
a
g

e
 

g
ro

u
p

s 
2

5
-2

7;
 3

5
-3

7
) 

 
R

e
la

ti
o

n
sh

ip
 (

0
/1

) 
M

ar
ri

e
d

 (
0
/1

) 
C

h
ild

le
ss

 (
0
/1

) 
 

M
o

d
e

l 1
 

M
o

d
e

l 2
 

M
o

d
e

l 3
 

M
o

d
e

l 4
 

M
o

d
e

l 5
 

M
o

d
e

l 6
 

M
o

d
e

l 7
 

M
o

d
e

l 8
 

M
o

d
e

l 9
 

N
o

n
-m

o
b

ile
 (

R
e
f.

: 
S

h
o

rt
-d

is
ta

n
ce

 c
o

m
m

u
te

rs
) 

.8
9

9
 

.8
5

3
 

.7
9

6
 

1
.3

0
1

* 
1

.0
7

0
 

.8
7

4
 

.6
9

3
**

.8
5

9
 

1
.0

7
8

 

M
e
d

iu
m

-d
is

ta
n

ce
 c

o
m

m
. 

(R
e
f.

: 
S

h
o

rt
-d

is
ta

n
ce

 
co

m
m

u
te

rs
) 

.8
3

3
* 

.9
7

3
 

.9
1

4
 

.9
8

3
 

1
.2

1
7

+
 

1
.1

8
4

 
1

.2
5

7
**

.8
9

3
 

.9
1

4
 

Lo
n

g
-d

is
ta

n
ce

 c
o

m
m

. 
(R

e
f.

: 
S

h
o

rt
-d

is
ta

n
ce

 
co

m
m

u
te

rs
) 

.9
1

8
 

1
.0

9
4

 
1

.0
0

4
 

.8
0

9
+
 

1
.1

2
0

 
1

.0
6

4
 

1
.3

4
0

* 
.8

1
1

 
.8

1
9

 

O
v
e
rn

ig
h

te
rs

 (
R

e
f.

: 
S

h
o

rt
-d

is
ta

n
ce

 c
o

m
m

u
te

rs
) 

.9
7

0
 

1
.0

5
2

 
1

.0
2

8
 

.7
9

6
* 

1
.0

7
2

 
1

.1
4

6
 

1
.4

4
5

**
.9

6
6

 
.9

7
7

 

N
o

n
-m

o
b

ile
*w

o
m

e
n

 
 

1
.1

4
2

 
1

.1
4

7
 

1
.5

0
3

+
 

1
.7

4
3

* 
.6

2
0

* 
.4

1
1

**
 

M
e
d

iu
m

-d
is

ta
n

ce
 c

o
m

m
.*

w
o

m
e

n
 

 
.7

0
4

* 
.7

7
3

 
.6

4
9

**
 

.8
2

8
 

2
.0

0
4

**
*

1
.9

1
8

**
 

Lo
n

g
-d

is
ta

n
ce

 c
o

m
m

.*
w

o
m

e
n

 
 

.6
4

6
+
 

.7
5

6
 

.4
6

6
**

 
.6

9
1

 
3

.2
4

9
**

*
2

.6
2

0
**

 

O
ve

rn
ig

h
te

r*
w

o
m

e
n

 
 

.7
8

5
 

.8
8

8
 

.3
6

0
**

*
.4

2
1

**
3

.7
2

6
**

*
2

.7
0

1
**

 

W
o

m
e
n

 (
R

e
f.

: 
m

e
n

) 
1

.3
3

3
**

*
1

.4
9

6
**

*
1

.3
7

1
**

.9
8

3
 

1
.1

7
5

* 
.9

5
6

 
.7

3
1

**
*

1
.2

1
3

+
 

A
g

e
 3

5
-3

7
 (

R
e
f.

: 
a
g

e
 2

5
-2

7
) 

 
2

.0
8

2
**

*
 

7
.2

4
5

**
*

.0
9

1
**

*

C
o

m
p

le
te

d
 s

e
co

n
d

ar
y
 s

ch
o

o
l (

R
e
f.

: 
e
le

m
e
n

ta
ry

) 
 

1
.1

8
5

+
 

 
.6

8
0

**
*

1
.6

0
3

**
*

U
p

p
e
r 

se
co

n
d

a
ry

 s
ch

o
o

l (
R

e
f.

: 
e

le
m

e
n

ta
ry

) 
 

1
.0

0
4

 
 

.5
6

4
**

*
2

.8
3

0
**

*

U
n

iv
e
rs

it
y
 d

e
g

re
e
 (

R
e
f.

: 
e
le

m
e
n

ta
ry

) 
 

1
.3

2
4

* 
 

.6
1

2
**

*
2

.6
3

7
**

*

IS
E

I 
(4

1
-6

5
) 

(R
e
f.

: 
IS

E
I 
1
6
-4

0
) 

 
1

.0
7

9
 

 
.9

1
6

 
1

.4
6

4
**

*

IS
E

I 
(6

6
-9

0
) 

(R
e
f.

: 
IS

E
I 
1
6
-4

0
) 

 
1

.2
8

6
* 

 
.9

2
5

 
1

.2
2

6
 

W
e
st

 (
R

e
f.

: 
E

a
st

) 
 

1
.1

6
1

+
 

 
1

.8
5

1
**

*
1

.9
7

8
**

*

F
u

ll-
ti

m
e
 (

R
e
f.

: 
p

a
rt

-t
im

e
) 

 
.8

9
8

 
 

.4
7

6
**

*
6

.5
5

1
**

*

N
a
g

e
lk

e
rk

e
 R

2
 

0
.0

0
7

 
0

.0
0

9
 

0
.0

5
5

 
0

.0
0

5
 

0
.0

1
5

 
0

.3
0

1
 

0
.0

1
2

 
0

.0
3

2
 

0
.4

3
9

 

S
o

u
rc

e
: 

p
ai

rf
am

 (
w

av
e
 1

, 
2
0
0
8
/2

0
0
9
);

 u
n

w
e
ig

h
te

d
 d

at
a;

 o
w

n
 c

al
cu

la
ti

o
n

s;
 b

in
ar

y
 l

o
g

is
ti

c 
re

g
re

ss
io

n
s;

 +
=

p
<

0
.1

0
; 

*=
p

<
0
.0

5
; 

**
=

p
<

0
.0

1
; 

**
*=

p
<

0
.0

0
1



•    Heiko Rüger, Michael Feldhaus, Katharina S. Becker, Monika Schlegel242

6, model 8), it can be stated that especially women exhibit the following correlation: 
Women (as compared to men) with more demanding forms of mobility are repre-
sented much more often among the group of persons who are childless. These 
effects of being childless remain unchanged even if reviewed based on additional 
covariates that also show extremely signifi cant effects. These analyses confi rm ex-
isting fi ndings that indicate that the mobility status of men is not to the same extent 
correlated to having or not having children. In contrast, highly mobile women show 
higher ratios of current childlessness when compared to the group of non-mobile 
women (Schneider et al. 2002b: 256; Limmer 2005; Schneider et al. 2008; Schnei-
der et al. 2009b; Meil 2010a/b).

The analyses presented here can be used to clearly illustrate that circular job 
mobility can represent a relevant factor within the scope of relationship and family 
developments – especially inasmuch as women are concerned. In it, we have iden-
tifi ed substantial effects for Long-distance commuters and particularly for Over-
nighters: while the effect on the partnership status is only minor, the likelihood of 
a person being married or being childless varies considerably depending on the 
mobility status. Highly mobile women, who often spend nights away from home 
for work reasons, are less likely to live in a highly institutionalised partnership (mar-
riage) and to be mothers. When controlling for additional relevant factors, these 
effects decrease, yet they remain signifi cant especially in the pairfam study that 
included a larger sampling size. The fi ndings, according to which circular job mobil-
ity appears to more strongly infl uence whether an individual plans to start a family 
than whether he or she lives with a partner or is married, may possibly indicate cu-
mulative effects. What is known is that in western Germany especially, highly insti-
tutionalised relationships are often seen as a prerequisite for a couple’s decision to 
have children. In addition, the (absent) fi ndings pertaining to a person’s relationship 
status might also point to the opposite effect, that is, it could be found that an exist-
ing relationship might be a reason why individuals are  more mobile (e.g. Schneider 
et al. 2008; Rüger/Becker 2011). Conversely, mobility may often be a reason for indi-
viduals to (unintentionally) remain single (Rüger/Ruppenthal 2010).

4 Discussion 

This article examined three central objectives. In a fi rst step, we developed a com-
mon indicator for circular job mobility patterns to be used in the two representative 
surveys ”Job Mobilities and Family Lives in Europe“ (JobMob) and ”Panel Analysis 
of Intimate Relationships and Family Dynamics“ (pairfam). Based on this common 
indicator, we were able to analyse the prevalence of a variety of mobility forms and 
their composition according to key socio-demographic criteria for both studies by 
means of a comparative analysis. In addition, we analysed correlations between 
the various forms of circular job mobility and partnership as well as family develop-
ment. 

Whereas a large number of analyses have been conducted in the fi eld of mobility 
and family research in response to the European comparative JobMob study, this is 



Circular job-related spatial mobility in Germany    • 243

not yet the case with regard to the data from the pairfam study. One particular fea-
ture of this article was its comparative approach, with the advantage being that the 
research results from the two studies have directly reinforced each other. In terms 
of the non-standardised constructs in social sciences, the opportunities available 
for comparison only exist to a minimal extent in different studies. Only very rarely 
are consolidated, comparative indicators used. However, comparative analyses are 
helpful as a means to review constructs in terms of their reliability and their gener-
alisation potential. One important objective of this article was to ensure reliability 
and to make generalisations about the concept of circular job mobility. 

As concerns the prevalence of individual forms of circular job mobility, overall 
there were only minor differences. About one out of two gainfully employed individ-
uals in Germany commutes short distances and almost one in fi ve commutes me-
dium distances. One out of seven gainfully employed individuals have to deal with 
even more extensive mobility arrangements, such as long-distance commuting or 
the need to frequently spend nights away from their permanent place of residence. 
This proves that the prevalence of more comprehensive forms of circular mobility is 
by no means a marginal phenomenon in our society. 

The correlations between mobility patterns and socio-demographic criteria in 
both studies are quite similar as well. For example, we have found a gender-specifi c 
distribution among the different forms of mobility and a correlation with education: 
persons with a lower level of education were more frequently Short-distance com-
muters, whereas individuals with a higher level of education exhibited longer com-
muting times and faced higher mobility demands. Regional and structural criteria 
played a role with regard to differences between eastern and western Germany. In 
the west, lower mobility demands were more common. By contrast, the labour mar-
ket places higher mobility demands on employees in the east. These differences are 
most likely the result of infrastructural differences and differing levels of ”mobility 
pressure“ on the labour market.  

Finally, on the basis of multivariate comparative analyses we were able to show 
that circular job mobility can have an impact on partnership and family development. 
Substantial effects were apparent in particular for the group of Long-distance com-
muters and Overnighters. However, this applies to a lesser extent to the likelihood 
of a person living with a partner. The assumption that singles especially are among 
those who deal with mobility issues to a greater extent could not be confi rmed in 
either of the studies. However, both studies produced comparative fi ndings indi-
cating that the transition to marriage or to the process of starting a family within 
existing partnerships does strongly correlate with circular job mobility. Therefore, 
we have successfully confi rmed that mobility plays an especially signifi cant role 
once the partnership-related opportunity structure increases in importance. This, in 
turn, may affect the opportunities available in the day-to-day structuring of the re-
lationship, the couple’s closeness of interaction, the amount of time they can spend 
together, and factors that can be assumed to be irreconcilable with high mobility 
requirements, such as the birth of a child. This is particularly the case when more 
extensive forms of organising become necessary, for example a couple’s ability to 
care for their children. To date, women are still more strongly affected by this than 



•    Heiko Rüger, Michael Feldhaus, Katharina S. Becker, Monika Schlegel244

men. Therefore, the empirical fi ndings can also be summed up as follows: If circular 
mobility plays a role in the context of personal relationships and family, then this is 
particularly the case for women. 

It should be emphasised that the comparative analyses presented here are 
cross-sectional. As a result, any causal explanations will clearly be limited. There-
fore, future studies conducted on the basis of longitudinal designs seem promising. 
For example, when do changes arise within a person’s life with regard to his or her 
mobility behaviour? What causal infl uence do the different forms of circular mobility 
have on the processes of institutionalising intimate relationships? How do inten-
tions (e.g. with regard to starting a family) and decisions relating to the dynamics of 
the relationship change when mobility patterns change? These research questions, 
which should be conducted with the help of longitudinal study designs, may in fact 
play a role in decrypting the specifi c adaptation and selection processes relating to 
an individual’s mobility behaviour and the associated steps and strategies a person 
takes to deal with partnerships and family-relevant aspects. Concerning the ques-
tion of how relationships are formed and institutionalised, the use of this longitu-
dinal data may allow us to pose the question much more clearly than ever before 
with regard to the correlation between the processes that determine the choice of a 
partner and the steps leading to institutionalising the relationship. Is mobility more 
often a barrier to partnerships or can it in fact be useful for a person’s efforts to 
deal with certain demands that only arise within a relationship, such as the ability to 
synchronise two professional careers? As concerns the correlation between mobil-
ity and the processes involved in starting a family, preliminary longitudinal analyses 
show that when a person stops commuting medium or long distances, it indicates 
that there is an increased likelihood that the person intends to start a family. In the 
future, a greater amount of attention should be paid to these kinds of longitudinal 
analyses. Both studies now offer the relevant empirical foundation.

Acknowledgements
The authors would like to thank the two anonymous reviewers for their valuable 
comments as well as Thomas Skora for his support in an earlier version of this ar-
ticle. 

References

Backhaus, Klaus et al. 2003: Multivariate Analysemethoden. Eine anwendungsorienti-
erte Einführung. Berlin u.a.: Springer [doi: 10.1007/978-3-642-14987-0].

Brüderl, Josef et al. 2010: pairfam Data Manual, Wave 1. Mannheim.

Castells, Manuel 2003 [1996-1999]: Das Informationszeitalter. 3 volumes. Wiesbaden: 
VS Verlag.

Collmer, Sabine 2005: Soldat, Familie und Mobilität: Neue Trends zur Lösung wider-
sprüchlicher Anforderungen. In: Kümmel, Gerhard (Eds.): Diener zweier Herren: Sol-
daten zwischen Bundeswehr und Familie. Frankfurt a. M.: Lang: 53-78.



Circular job-related spatial mobility in Germany    • 245

Feldhaus, Michael; Schlegel, Monika 2011: Einfl üsse der Partnerschaftsdynamik auf 
den Übergang zur Kohabitation und zur Heirat. In: Brüderl, Josef; Castiglioni, Laura; 
Schumann, Nina (Eds.): Partnerschaft, Fertilität und intergenerationale Beziehungen. 
Würzburg: Ergon: 73-105.

Grau, Andreas 2009: Pendler: Die Mehrheit nimmt weiter das Auto. STATMagazin. Wi-
esbaden: Destatis.

Haas, Anette; Hamann, Silke 2008: Pendeln – ein zunehmender Trend, vor allem bei 
Hochqualifi zierten. IAB-Kurzbericht 6/2008. Nürnberg. 

Huinink, Johannes et al. 2011: Panel Analysis of Intimate Relationships and Family Dy-
namics (pairfam): Conceptual Framework and Design. In: Zeitschrift für Familienfor-
schung 23,1: 77-101.

Huinink, Johannes; Feldhaus, Michael 2008: Beziehungs- und Familienentwicklung – 
eine konzeptionelle Einführung in ein Forschungsprogramm. In: Feldhaus, Michael; 
Huinink, Johannes (Eds.): Neuere Entwicklungen in der Beziehungs- und Familien-
forschung. Vorstudien zum Beziehungs- und Familienentwicklungspanel (PAIRFAM). 
Würzburg: Ergon: 13-45.

Huinink, Johannes; Wagner, Michael 1989: Regionale Lebensbedingungen, Migration 
und Familienbildung. In: Kölner Zeitschrift für Soziologie und Sozialpsychologie 4,4: 
669-689.

Huynen, Philippe; Hubert, Michel; Lück, Detlev 2010: Research Design. In: Schneider, 
Norbert F.; Collet, Beate (Eds.): Mobile Living Across Europe II. Causes and Conse-
quences of Job-Related Spatial Mobility in Cross-National Comparison. Opladen/
Farmington Hills: Barbara Budrich Publishers: 25-36.

Huynen, Philippe et al. 2008: Survey Design and Methods. In: Schneider, Norbert F.; 
Meil, Gerardo (Eds.): Mobile Living Across Europe I. Relevance and Diversity of Job-
Related Spatial Mobility in Six European Countries. Opladen/Farmington Hills: Bar-
bara Budrich Publishers: 47-63.

Jaccard, James 2001: Interaction Effects in Logistic Regression. Quantitative Applica-
tions in the Social Sciences 135. Thousand Oaks: Sage [doi: 10.4135/9781412984515. 

Kley, Stefanie 2009: Migration im Lebensverlauf. Wiesbaden: VS Verlag.

Kulu, Hill 2005: Migration and Fertility: Competing Hypotheses Re-examined. In: Euro-
pean Journal of Population 21,1: 51–87 [doi: 10.1007/s10680-005-3581-8].

Limmer, Ruth 2005: Berufsmobilität und Familie in Deutschland. In: Zeitschrift für Fami-
lienforschung 17,2: 8-26.

Limmer, Ruth; Rüger, Heiko 2010: Job Mobilities and Quality of Life. In: Schneider, Nor-
bert F.; Collet, Beate (Eds.): Mobile Living Across Europe II. Causes and Consequences 
of Job-Related Spatial Mobility in Cross-National Comparison. Opladen/Farmington 
Hills: Barbara Budrich Publishers: 263-288.

Limmer, Ruth; Schneider, Norbert F. 2008: Studying Job-Related Spatial Mobility in Eu-
rope. In: Schneider, Norbert F.; Meil, Gerardo (Eds.): Mobile Living Across Europe I. 
Relevance and Diversity of Job-Related Spatial Mobility in Six European Countries. 
Opladen/Farmington Hills: Barbara Budrich Publishers: 13-46.

Lück, Detlev; Ruppenthal, Silvia 2010: Insights into Mobile Living: Spread, Appearanc-
es and Characteristics. In: Schneider, Norbert F.; Collet, Beate (Eds.): Mobile Living 
Across Europe II. Causes and Consequences of Job-Related Spatial Mobility in Cross-
National Comparison. Opladen/Farmington Hills: Barbara Budrich Publishers: 37-68.



•    Heiko Rüger, Michael Feldhaus, Katharina S. Becker, Monika Schlegel246

Lück, Detlev; Schneider, Norbert F. 2010: Introduction to the Special Issue on Mobility 
and Familiy. Increasing Job Mobility – Changing Family Lives. In: Zeitschrift für Fami-
lienforschung 22,2: 135-148.

Meil, Gerardo 2010a: Geographic Job Mobility and Parenthood Decisions. In: Zeitschrift 
für Familienforschung 22,2: 171-196.

Meil, Gerardo 2010b: Job Mobility and Family Life. In: Schneider, Norbert F.; Collet, 
Beate (Eds.): Mobile Living Across Europe II. Causes and Consequences of Job-Relat-
ed Spatial Mobility in Cross-National Comparison. Opladen/Farmington Hills: Barbara 
Budrich Publishers: 215-236.

Nolle, Tanja 2005: Mobile Berufe. Eine Untersuchung der Arbeitsbedingungen und der 
Ernährung im Hinblick auf die Gesundheit. Dortmund: Universität Dortmund.

Peuckert, Rüdiger 2008: Familienformen im sozialen Wandel. Wiesbaden: VS Verlag.

Raithel, Jürgen 2008: Quantitative Forschung. Ein Praxiskurs. Wiesbaden: VS Verlag.

Reuschke, Darja 2010: Job-induced Commuting between Two Residences – Character-
istics of a Multilocational Living Arrangement in the Late Modernity. In: Compara-
tive Population Studies – Zeitschrift für Bevölkerungswissenschaft 35,1: 107-134 [doi: 
10.4232/10.CPoS-2010-04en].

Rüger, Heiko; Becker, Katharina 2011: Berufsmobilität, Geschlecht und Lebensform. 
Berufsbedingte räumliche Mobilität und ihre Folgen für die Vereinbarkeit von Fami-
lie und Beruf und partnerschaftliche Arbeitsteilung. In: Klammer, Ute; Motz, Markus 
(Eds.): Neue Wege – Gleiche Chancen. Expertisen zum Ersten Gleichstellungsbericht 
der Bundesregierung. Wiesbaden: VS Verlag: 363-396.

Rüger, Heiko; Ruppenthal, Silvia 2010: Advantages and Disadvantages of Job-related 
Spatial Mobility. In: Schneider, Norbert F.; Collet, Beate (Eds.): Mobile Living Across 
Europe II. Causes and Consequences of Job-Related Spatial Mobility in Cross-National 
Comparison. Opladen/Farmington Hills: Barbara Budrich Publishers: 69-94.

Ruppenthal, Silvia; Lück, Detlev 2009: Jeder fünfte Erwerbstätige ist aus berufl ichen 
Gründen mobil. Berufbedingte räumliche Mobilität im Vergleich. In: ISI (Informations-
dienst soziale Indikatoren) 42: 1-5.

Schier, Michaela 2009: Räumliche Entgrenzung von Arbeit und Familie. Die Herstellung 
von Familie unter Bedingungen von Multilokalität. In: Informationen zur Raumentwick-
lung 1-2/2009: 55-66.

Schneider, Norbert F. 2005: Leben an zwei Orten. Die Folgen berufl icher Mobilität für 
Familie und Partnerschaft. In: Mischau, Anina; Oechsle, Mechtild (Eds.): Arbeitszeit – 
Familienzeit – Lebenszeit: Verlieren wir die Balance? (Zeitschrift für Familienforschung 
Sonderheft 5). Wiesbaden: VS Verlag: 110-126.

Schneider, Norbert F.; Lück, Detlev; Ruppenthal, Silvia; Rüger, Heiko 2011: Code 
Book for the Job Mobilities and Family Lives Data Set First Wave. JobMob and Fam-
Lives Working Paper (JFW), 11-02, available on: www.jobmob-and-famlives.eu or 
www.gesis.org. 

Schneider, Norbert F.; Collet, Beate (Eds.) 2010: Mobile Living Across Europe II. Causes 
and Consequences of Job-Related Spatial Mobility in Cross-National Comparison. Op-
laden/Farmington Hills: Barbara Budrich Publishers.

Schneider, Norbert F.; Ruppenthal, Silvia; Lück, Detlev 2009a: Beruf, Mobilität und 
Familie. In: Burkart, Günter (Eds.): Zukunft der Familie. Prognosen und Szenarien. 
(Zeitschrift für Familienforschung Sonderheft 6). Opladen/Farmington Hills: Barbara 
Budrich Publishers: 111-136. 



Circular job-related spatial mobility in Germany    • 247

Schneider, Norbert F.; Rüger, Heiko; Münster, Eva 2009b: Berufsbedingte räumliche Mo-
bilität in Deutschland. Formen, Verbreitung und Folgen für Gesundheit, Wohlbefi nden 
und Familienleben. In: Arbeitsmedizin Sozialmedizin Umweltmedizin 44,7: 400-409.

Schneider, Norbert F.; Meil, Gerardo (Eds.) 2008: Mobile Living Across Europe I. Rel-
evance and Diversity of Job-Related Spatial Mobility in Six European Countries. Op-
laden/Farmington Hills: Barbara Budrich Publishers.

Schneider, Norbert F. et al. 2008: Germany – A Country of Locally Attached but Highly 
Mobile People. In: Schneider, Norbert F.; Meil, Gerardo (Eds.): Mobile Living Across 
Europe I. Relevance and Diversity of Job-Related Spatial Mobility in Six European 
Countries. Opladen/Farmington Hills: Barbara Budrich Publishers: 105-147.

Schneider, Norbert F.; Limmer, Ruth; Ruckdeschel, Kerstin 2002a: Mobil, fl exibel, ge-
bunden. Familie und Beruf in der mobilen Gesellschaft. Frankfurt am Main/New York: 
Campus Verlag.

Schneider Norbert F.; Limmer Ruth; Ruckdeschel, Kerstin 2002b: Berufsmobilität und 
Lebensform. Schriftenreihe des Bundesministeriums für Familie, Senioren, Frauen 
und Jugend 208. Stuttgart: Kohlhammer.

Sennett, Richard 1998: Der fl exible Mensch. Die Kultur des neuen Kapitalismus. Berlin: 
Berliner Taschenbuch Verlag.

Statistisches Bundesamt 2005: Leben und Arbeiten in Deutschland. Mikrozensus 2004. 
Wiesbaden.

Stutzer, Alois; Frey, Bruno S. 2008: Stress that Doesn’t Pay. The Commuting Para-
dox. In: The Scandinavian Journal of Economics 110,2: 339-366 [doi: 10.1111/j.1467-
9442.2008.00542.x].

Thomas, William; Znaniecki, Florian 1996 [1918-1920]: The Polish Peasant in Europe and 
America: A Classic Work in Immigration History. o.O.: University of Illinois Press.

Urry, John 2007: Mobilities. Cambridge/Malden: Polity Press.

Wagner, Michael 1989: Räumliche Mobilität im Lebensverlauf. Eine empirische Untersu-
chung sozialer Bedingungen der Migration. Stuttgart: Ferdinand Enke Verlag.

Weichhart, Peter 2009: Multilokalität – Konzepte, Theoriebezüge und Forschungsfragen. 
In: Informationen zur Raumentwicklung 1-2/2009: 1-14. 

Wiethölter, Doris; Bogai, Dieter; Carstensen, Janette 2009: Pendlerbericht Berlin-Bran-
denburg 2009. In: IAB-Regional 3/2010.

Zelinsky, Wilbur 1971: The Hypothesis of the Mobility Transition. In: Geographical Re-
view 61,2: 219–249.



•    Heiko Rüger, Michael Feldhaus, Katharina S. Becker, Monika Schlegel248

Translated from the original text by the Federal Institute for Population Research, for information 
only. The reviewed and author’s authorised original article in German is available under the title 
“Zirkuläre berufsbezogene Mobilität in Deutschland: Vergleichende Analysen mit zwei repräsenta-
tiven Surveys zu Formen, Verbreitung und Relevanz im Kontext der Partnerschafts- und Fami-
lienentwicklung”, DOI 10.4232/10.CPoS-2011-05de or URN urn:nbn:de:bib-cpos-2011-05de1, at 
http://www.comparativepopulationstudies.de.

Date of submission: 06.06.2011   Date of Acceptance: 21.11.2011

Heiko Rüger ( ), Katharina S. Becker. Federal Institute for Population Research, 
Wiesbaden, Germany. 
E-Mail: heiko.rueger@destatis.de, katharina.becker@destatis.de 
URL: www.bib-demografi e.de

Dr. Michael Feldhaus. Universität Bremen, Institut für empirische und angewandte 
Soziologie (EMPAS), Bremen, Germany. E-Mail: feldhaus@empas.uni-bremen.de
URL: www.empas.uni-bremen.de

Dr. Monika Schlegel. University of Vechta, Institute for Social Work, Education and Sport 
Sciences, Vechta, Germany. E-Mail: monika.schlegel@uni-vechta.de
URL: www.uni-vechta.de



Comparative Population Studies – Zeitschrift für Bevölkerungswissenschaft

www.comparativepopulationstudies.de

ISSN: 1869-8980 (Print) – 1869-8999 (Internet)

© Federal Institute for Population Research 2011 – All rights reserved

Published by / Herausgegeben von
Prof. Dr. Norbert F. Schneider

Layout and print: Federal Institute for 
Population Research, Wiesbaden 
(Germany)

Managing Editor / Redaktion
Frank Swiaczny

Copy Editor / Schlussredaktion
Dr. Evelyn Grünheid

Scientifi c Advisory Board / 
Wissenschaftlicher Beirat

Jürgen Dorbritz (Wiesbaden) 
Paul Gans (Mannheim) 
Johannes Huinink (Bremen) 
Marc Luy (Wien) 
Clara H. Mulder (Groningen) 
Notburga Ott (Bochum) 
Peter Preisendörfer (Mainz)

Board of Reviewers / Gutachterbeirat

Martin Abraham (Erlangen)
Laura Bernardi (Lausanne)
Hansjörg Bucher (Bonn)
Claudia Diehl (Göttingen)
Andreas Diekmann (Zürich)
Gabriele Doblhammer-Reiter (Rostock)
Henriette Engelhardt-Wölfl er (Bamberg)
E.-Jürgen Flöthmann (Bielefeld)
Alexia Fürnkranz-Prskawetz (Wien)
Beat Fux (Zürich)
Joshua Goldstein (Rostock)
Karsten Hank (Köln)
Sonja Haug (Regensburg)
Franz-Josef Kemper (Berlin)
Michaela Kreyenfeld (Rostock)
Aart C. Liefbroer (Den Haag)
Kurt Lüscher (Konstanz)
Dimiter Philipov (Wien)
Tomáš Sobotka (Wien)
Heike Trappe (Rostock)