Wach e , Isabelle; Holz-Rau, Ch is ian
A icle — Published Ve sion
Gende di e ences in wo k- ela ed high mobili y
di e en ia ed by pa ne ship and pa en hood s a us
T anspo a ion
P o ided in Coope a ion wi h:
Sp inge Na u e
Sugges ed Ci a ion: Wach e , Isabelle; Holz-Rau, Ch is ian (2021) : Gende di e ences in wo k- ela ed
high mobili y di e en ia ed by pa ne ship and pa en hood s a us, T anspo a ion, ISSN 1572-9435,
Sp inge US, New Yo k, NY, Vol. 49, Iss. 6, pp. 1737-1764,
h ps://doi.o g/10.1007/s11116-021-10226-z
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/287069
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h ps://c ea i ecommons.o g/licenses/by/4.0/
Vol.:(0123456789)
T anspo a ion (2022) 49:1737–1764
h ps://doi.o g/10.1007/s11116-021-10226-z
1 3
Gende di e ences inwo k‑ ela ed high mobili y
di e en ia ed bypa ne ship andpa en hood s a us
IsabelleWach e 1 · Ch is ianHolz‑Rau1
Accep ed: 24 Augus 2021 / Published online: 25 Sep embe 2021
© The Au ho (s) 2021
Abs ac
The income si ua ion and he di ision o labo in households, which a e closely ela ed o
occupa ional mobili y, a e cen al aspec s o he deba e on gende equali y. Women ha e
sho e commu ing imes and dis ances han men and spend ewe nigh s away om hei
main place o esidence o wo k- ela ed easons. Va ious s udies a ibu e hese gende di -
e ences o a gende ed di ision o labo and he associa ed g ea e in ol emen o women
in household asks and childca e. Consequen ly, s udies in es iga ing hese gende di e -
ences ocus p ima ily on employees in ela ionships and he associa ed in a-couple in e ac-
ions, while li le a en ion is paid o singles. Based on he Ge man Family Panel (pai am)
his esea ch aims o b oaden he scope o in e p e a ion and examines gende di e ences
in wo k- ela ed high mobili y among employees in pa ne ships wi h and wi hou child en
and among singles. Logis ic eg ession models including gende in e ac ion e ms show
ha gende di e ences exis no only among employees wi h pa ne s (and child en), bu
also among singles. The esul s highligh ha gende di e ences in high mobili y a e due
o ac o s ela ed o ela ionships and pa en hood, as well as om o he ac o s. Gende
di e ences in high mobili y a e hus no me ely he esul o nego ia ion p ocesses o o
(pa ia chal) powe s uc u es in ela ionships and gende ed labo di ision. They a e also
ela ed o gende ed occupa ional seg ega ion and economic dispa i ies and in e nalized
gende p e e ences ha a e independen o pa ne ship and pa en hood.
Keywo ds Long-dis ance commu ing· O e nigh ing· Gende · Pa en hood· In e ac ion
e ms
* Isabelle Wach e
isabelle.w[email p o ec ed]
Ch is ian Holz-Rau
[email p o ec ed]
1 Depa men o Spa ial Planning, T anspo Resea ch G oup, TU Do mund Uni e si y,
44227Do mund, Ge many
1738
T anspo a ion (2022) 49:1737–1764
1 3
In oduc ion
The income si ua ion and he di ision o labo in households a e key aspec s o he deba e
on gende equali y. Bo h a e closely linked o wo k- ela ed mobili y, which connec s he
p o essional and domes ic sphe es. P e ious esea ch has consis en ly shown ha women
ha e sho e commu es han men in e ms o dis ance (C ane 2007; Sandow and Wes-
in 2010; Axisa e al. 2012) and ime (Eu opean Communi ies 2004; McQuaid and Chen
2012). Fu he mo e, women spend ewe nigh s away om hei main esidence due o
wo k (Colle and Daube 2010; Reuschke 2010a; Rüge e al. 2011). T adi ional gende
oles, whe e men a e he wage ea ne s and women a e esponsible o household asks and
childca e, a e conside ed one o he main d i e s o his gende di e ence. Acco dingly,
he majo i y o esea ch ocuses on couples (wi h child en) and associa ed in a-household
decisions (Plau 2006; an de Klis and Mulde 2008; Gimenez-Nadal and Molina 2016).
Only a ew s udies add ess hese gende di e ences among singles (households). They
mos ly indica e smalle gende di e ences among singles (Mauch and Taylo 1997; Chid-
amba am and Scheine 2020; Hu 2020).
Based on ha , his esea ch examines gende di e ences in wo k- ela ed high mobili y
(daily long-dis ance commu e s and o e nigh e s) among pe sons in ela ionships wi h and
wi hou child en and among singles based on he Ge man Family Panel (pai am). Addi-
ionally, we in es iga e how indi idual and household- ela ed a ibu es in e ac wi h gen-
de in e ms o high mobili y. This app oach aims o iden i y how ac o s ela ed o pa ne -
ship and pa en hood, as well as o he ac o s, a e linked wi h wo k- ela ed high mobili y
and hus expand he scope o in e p e a ion o gende di e ences in high mobili y.
In henex sec ion, we summa ize he heo e ical pe spec i es and p e ious esea ch on
gende di e ences in wo k- ela ed mobili y.This is ollowed by he “Da a and me hodol-
ogy”sec ion and he “Resul s”sec ion. We conclude wi h  he“Discussion and conclu-
sions” sec ion, whe e we also poin ou limi a ions and u he esea ch needs.
Theo e ical backg ound andli e a u e e iew
High mobili y: long‑dis ance commu ing ando e nigh ing
De ini ions o daily long-dis ance commu ing (LDC)1 a y depending on s udy a ea and
ocus. Dis ance-based de ini ions ha e a conside able ange, e.g. 17km pe way in Can-
ada (Maoh and Tang 2012), 30km in Sweden (Sandow and Wes in 2010) and 50km in
Ge many (S a is isches Bundesam and Wissenscha szen um Be lin ü Sozial o schung
2018). De ini ions based on a el ime o en se he h eshold a 60min one-way o a
an o e all commu e ime o 120min2 (Schneide e al. 2001, Schneide and Meil 2008,
Vincen -Geslin and Ra ale 2016). While he e a e many s udies on LDC and commu -
ing imes and dis ances in gene al, he e is less esea ch on o e nigh e s. The s udy Job
Mobili ies and Family Li es in Eu ope (JobMob) (Schneide and Meil 2008, Schneide and
Colle 2010) and u he wo k pa ly based on he JobMob da a (Rüge e al. 2011; Ra ale
1 Some imes also called ex eme commu ing (Maoh and Tang 2012; Ra ale e al. 2015).
2 Al hough he e m "long-du a ion commu e " would be mo e app op ia e, we use he e m “long-dis ance
commu e ” o easons o consis ency.
1739
T anspo a ion (2022) 49:1737–1764
1 3
e al. 2014; Vi y and Kau mann 2015) examine se e al o ms o job- ela ed spa ial mobil-
i y including LDC and o e nigh ing.
O e nigh e s o en spend nigh s away om home due o wo k- ela ed easons, wi h he
h eshold usually se a 60 nigh s pe yea (Huynen e al. 2008; Rüge e al. 2011; Ra ale
e al. 2015). O e nigh ing subsumes a a ie y o mobili y o ms and hus includes shu les,
who a e also known as weekly o mon hly long-dis ance commu e s, and a i-mobiles,
who a el equen ly and a i egula in e als o changing wo k loca ions. Schneide and
Meil (2008) and Vi y and Kau mann (2015) also classi y people in long-dis ance ela ion-
ships as o e nigh e s. In con as o couples li ing apa oge he (LAT), hei easons o
no sha ing a common household wi h hei pa ne s a e p o essional a he han p i a e.
Al hough LDC, shu les, and a i-mobiles di e in he aspec s men ioned, hey a e simi-
la in impo an aspec s o he p esen s udy. This includes ha all h ee o ms o mobil-
i y cons i u e a s a egy o econcile he own pe sonal, social and p o essional li e and—i
applicable—addi ionally ha o he pa ne and child en (Limme and Schneide 2008;
And eo i e al. 2013). All h ee ypes o mobili y a e e y ime-in ensi e and he e o e
ha e a high impac on o he li e domains as hey educe he a ailable ime o hem (Lim-
me and Schneide 2008; Lück and Ruppen hal 2010; Ra ale e al. 2015). Thus, hey “a e
he esul o space and ime ade-o s be ween pe sonal and p o essional li es” (Ra ale
e al. 2015). Mo eo e , he e a e gende di e ences in each o he h ee ypes o high-
mobili y. These gende di e ences, which a e he ocus o ou s udy, a e explained by he
same heo ies ha we a e going o discuss in he ollowing sec ion. Fo hese easons, like
Vi y and Kau mann (2015), we use he e m high mobili y o summa ize LDC, shu les and
a i-mobiles.
High mobili y andgende di e ences
Despi e endencies o con e gence, analyses show ha women commu e sho e dis ances
and imes han men (C ane 2007; Kon ad 2016). Fu he mo e, hey a e mo e a ely o e -
nigh e s (G een e al. 1999; Colle and Daube 2010; Reuschke 2010a; Rüge e al. 2011).
Explana ions o hese gende di e ences exis in a ious scien i ic disciplines. In he ol-
lowing, we p esen he mos common hypo heses and explana ions.
Gende oles and he esul ing gende -speci ic di ision o labo a e conside ed one o
he main easons o gende di e ences in a el and commu ing beha io . The g ea e
in ol emen o women in household asks, including childca e, limi s women in ime and
space, which mani es s i sel in sho e commu es and less equen o e nigh ing compa ed
o men. This is also known as he Household Responsibili y Hypo hesis (HRH) (Hanson
and Johns on 1985; Johns on-Anumonwo 1992; Tu ne and Niemeie 1997; Sil ei a Ne o
e al. 2015; Gimenez-Nadal and Molina 2016). Despi e ising emale employmen a es,
social gende oles emain la gely unchanged (Col ane 2000; Wa en 2003). Acco dingly,
women unde ake mo e household-suppo and child-se ing ips han men (Schwanen
2007; Liu e al. 2012; Fan 2017). In line wi h he HRH, Chidamba am and Scheine (2020)
ind ha gende di e ences in commu ing in dual-ea ne households a e lowe i he male
pa ne spends mo e ime on unpaid wo k. Ne e heless, he majo i y o s udies examine
gende di e ences among he e osexual couples. Sma e al. (2017) ind ha gende di e -
ences also exis in same sex couples wi h espec o labo di ision and a el, bu hey a e
less p onounced.
O he explana ions ocus on occupa ional seg ega ion and economic dispa i-
ies. Female-domina ed occupa ions a e mo e e enly dis ibu ed in space han
1740
T anspo a ion (2022) 49:1737–1764
1 3
male-domina ed jobs, which a o s sho commu es (Hanson and Johns on 1985;
Singell and Lillydahl 1986; Hanson and P a 1988). Mo eo e , a ious s udies ha e
ound a posi i e associa ion be ween educa ion and income and high mobili y p ac ices
(Schwanen e al. 2004; Lück and Ruppen hal 2010; Sandow 2014). Full- ime employees
also commu e longe han pa - ime employees (Zolnik 2010; Axisa e al. 2012; Maoh
and Tang 2012) and a e disp opo iona ely ep esen ed among o e nigh e s (G een
e al. 1999; Bonne and Colle 2010; Reuschke 2010b). Howe e , women gene ally ha e
lowe incomes and a e mo e likely o wo k pa - ime, wi h he di e ences in Ge many
being especially la ge in an EU compa ison (Eu opean Ins i u e o Gende Equali y
2014; Boll and Lagemann 2018), which can cause gende di e ences in high mobili y.
Acco dingly, Madden’s esul s (1981) indica e ha gende balance in e ms o wo king
hou s and wages would lead o equal o e en longe commu es o women. O he s udies
show ha al hough gende di e ences dec ease when income and occupa ion a e con-
olled o , hey s ill emain (Go don e al. 1989; Rüge and Sulak 2017).
Income is also an impo an pa ame e in amily and household economics (Mince
1978; Becke 1991). Mince (1978) conside s mig a ion (e.g. esiden ial eloca ion)
decisions based on he o al (in)ma e ial bene i s and cos s o all household membe s.
I o al mig a ion bene i s exceed o al mig a ion cos s, he household eloca es. I he
cos s ou weigh he bene i s, e.g. i one pa ne ’s income gain om a mig a ion can-
no (o e )compensa e he loss o income o he o he pa ne , he household does no
eloca e and high mobili y becomes mo e ele an . Thus, he decision-making p ocess
in households ega ding eloca ion and high mobili y is mo e complex in mul i-pe son
households. High mobili y is some imes he only way o combine wo ca ee s and a
common place o esidence (in he p e e ed esiden ial a ea) (G een 1997; Vincen -
Geslin and Ra ale 2016). Nisic (2010) poin s ou ha indi idual e o s (commu ing/
(un)paid wo k) and bene i s ela ed o household decisions a e une enly dis ibu ed
among household membe s. Thus, a household’s consensus is he esul o nego ia ion
p ocesses, whe eby he ela i e posi ion o powe in pa ne ships de e mines which
in e es s p e ail (Blau 1986). Empi ical s udies o en measu e his posi ion o powe
using he a io o incomes (Nisic 2010). As men ioned abo e, women in gene al ha e
lowe incomes and hus weake ba gaining posi ions han hei pa ne s.
Feminis heo ies ace gende di e ences back o pa ia chal powe ela ions, which
shape gende oles and ela ionships. Pa ia chal powe ela ions hold women esponsi-
ble o household asks, hus es ic ing hei eedom and p omo ing gende inequali y
(Fens e make and Wes 2002). This gende hie a chy p i ileges men (Wood and Eagly
2002) and se es as an o ien a ion amewo k o (household-in e nal) decisions. Thus,
such powe ela ions exceed he economic di e ences desc ibed abo e (Ha mann 1976;
Walby 1994).
This is in s a k con as o o he heo ies, which Fan (2017) summa izes as heo ies
o in e nalized gende di e ences. These heo ies explain gende di e ences in gen-
de -speci ic pe sonali ies, a i udes and p e e ences (Hakim 2000). Empi ical e idence
indica es o example ha women ha e g ea e en i onmen al conce ns and esponsi-
bili y (Da idson and F eudenbu g 1996) and a e mo e inclined o dec ease hei ca
use han men (Ma hies e al. 2002; Polk 2004), which could explain he lowe wo k-
ela ed mobili y o women. Howe e , hese gende di e ences in a i udes and p e e -
ences could also esul om gende -speci ic socializa ion. Mo eo e , Reuschke and
Hous on (2020) ind, a e con olling o household esponsibili y, gende di e ences
in commu ing ime among employees, bu no among sel -employed pe sons who a e

1741
T anspo a ion (2022) 49:1737–1764
1 3
less a ec ed by labo ma ke cons ain s. They he e o e conclude ha labo ma ke con-
s ain s play a mo e impo an ole han gende p e e ences.
Mos o he hypo heses and explana ions o gende di e ences in high mobili y p e-
sen ed indica e some disad an ages/cons ain s o women. Acco ding o he hypo hesis
o pa ia chal powe ela ions, p esc ibed gende oles, and economic inequali ies and he
o en ela ed lowe ba gaining powe in he household nego ia ion p ocess, women a e p i-
ma ily esponsible o domes ic labo , al hough his may no co espond o hei p e e -
ences. This leads o empo a y and o spa ial cons ain s o women, which in u n limi s
hei mobili y, ac i i y spaces, access o labo ma ke s, and economic independence. How-
e e , he e a e also nume ous bu dens and disad an ages associa ed wi h high mobili y.
Rüge and Ruppen hal (2010) and Sandow e al. (2014) discuss he disad an ages o LDC
and o e nigh ing in de ail, we he e o e only b ie ly add ess he disad an ages o high
mobili y p ac ices. High mobili y esul s in inc eased le els o s ess (No aco e al. 1991;
Go holmsede e al. 2009). Mo eo e , Sandow e al. (2014) ind a highe mo ali y isk o
women who ha e expe ienced LDC han o women wi h sho e commu ing dis ances,
bu hey do no ind his associa ion o men. Addi ionally, high mobili y p ac ices sho en
he ime a ailable o o he ac i i ies, esul ing in unheal hy habi s such as less sleep and
exe cise o unheal hy ea ing habi s, which o e ime can cause heal h issues (Ch is ian
2012). Longe commu ing imes a e also associa ed wi h a highe isk o wo k-li e con lic s
(Hämmig e al. 2009), especially o women (Jansen e al. 2003). Acco ding o he JobMob
s udy, o e nigh e s in pa icula pe cei e a lack o ime o hei pa ne and child en as a
disad an age o hei high mobili y (Rüge and Ruppen hal 2010). Due o he disad an-
ages men ioned abo e and he nega i e en i onmen al impac o high mobili y, we ques-
ion whe he i is desi able o women o be as highly mobile as men. Fo his eason, we
a oid he no ma i e e m “gende gap” and p e e “gende di e ences”.
Gende di e ences inhigh mobili y in hecon ex o  ela ionship s a us
andhousehold
Mos o he explana ions and hypo heses e iewed in he p e ious sec ion explain he cause
o gende di e ences in high mobili y in e ms o di ision o labo , which esul s om gen-
de oles, nego ia ion p ocesses and/o in pa ia chal powe ela ions. I is he e o e likely
ha gende di e ences in high mobili y a e ela ed o pa ne ship and pa en hood.
Female long dis ance commu e s, shu les and a i-mobiles a e less likely o be in a
pa ne ship o o be ma ied han bo h hei male coun e pa s and women wi hou high
mobili y p ac ices (Rüge e al. 2011). Acco dingly, ma ied women ha e sho e com-
mu es han men and o he women wi hou child en (P es on e al. 1993; Lee and McDon-
ald 2003; P es on and McLa e y 2016). In con as , he e is no clea associa ion be ween
ela ionship s a us and high mobili y o men (Rüge e al. 2011). Some s udies e en ind
an inc ease in commu ing dis ance and ime wi h ma iage o men (P es on and McLa -
e y 2016). Consis en ly, gende di e ence in commu ing ends o be smalle among one-
adul households wi hou child en han in o he household ypes (Mauch and Taylo 1997;
Hu 2020). P es on and McLa e y (2016) do no iden i y any gende di e ences in com-
mu ing ime among unma ied pe sons wi hou child en. Zolnik (2010) inds no co ela ion
be ween ela ionship s a us and commu e ime. Fan (2017) also concludes ha pa en hood
a he han he pu e p esence o a pa ne is associa ed wi h la ge gende di e ences in
commu ing imes.
1742
T anspo a ion (2022) 49:1737–1764
1 3
Women who p ac ice long-dis ance commu ing o o e nigh ing a e less likely o
ha e child en compa ed o o he women and men (Schneide and Limme 2008; Meil
2010; Reuschke 2010b; Rüge e al. 2011; Rüge and Sulak 2017), and mo he s com-
mu e less han o he women (P es on e al. 1993; Sandow 2008) as well as men in
gene al (Schwanen e al. 2004). Fo men, on he o he hand, he nega i e co ela ion
be ween pa en hood and high mobili y is lowe (Sandow 2008, Rüge and Sulak 2017)
o does no exis a all (Rüge e al. 2011). Hu (2020) di e en ia es be ween wo-adul
households wi h one and wo wo ke s and inds ha ha ing child en gene ally does no
change gende di e ences in one-wo ke households bu inc eases gende di e ences in
wo-wo ke households. Howe e , Johns on-Anumonwo (1992) a gues ha he numbe
o wo ke s in a household a he han he p esence o child en is ela ed o gende di e -
ences in commu ing, and inds ha he commu ing dis ances o men and women di e
mo e in dual-wo ke households han in single-wo ke households. In con as , mo e
ecen s udies (Sul ana 2005; Fan 2017) ind no associa ion be ween b eadwinne s a us
and gende di e ences in commu e ime.
Rega ding he age o child en, Sandow (2008) inds a nega i e associa ion be ween
child en unde six and commu ing dis ance o bo h women and men, while child en o e
six a e nega i ely associa ed wi h commu ing dis ance only o women. Hjo hol (2000)
also iden i ies ha women wi h p eschool-age child en in pa icula wo k close o home.
Reuschke and Hous on (2020) ind he opposi e wi h espec o commu e ime. In hei
s udy, women wi h he younges child unde i e yea s commu e signi ican ly longe han
women wi hou dependen child en. Howe e , among pa en s, he age o he child en has
no signi ican e ec on commu e ime. Gimenez-Nadal and Molina (2016) also ind ha
age o child en does no appea o be a ele an ac o in commu e imes o men wi h chil-
d en, while mo he s wi h child en unde 5 yea s and 5 o 12 yea s end o commu e sligh ly
longe .
As no ed abo e, nume ous s udies con i m gende di e ences in high mobili y. How-
e e , mul i a ia e analyses usually include only he main e ec s o explana o y a iables.
Sepa a e eg essions o men and women a e o en used o iden i y gende -speci ic asso-
cia ions be ween independen a iables and high mobili y o commu ing in gene al (San-
dow 2008, Rüge and Sulak 2017). In e ac ions wi h gende a e a ely pa o mul i a ia e
analyses, al hough hey es whe he gende mode a es he e ec o independen a iables
on high mobili y. The ew analyses ha conside hese in e ac ions a e usually limi ed o
ce ain a iables (e.g. Reuschke 2010a; Rüge e al. 2011; Hu 2020). Mo eo e , mos s ud-
ies ocus on high mobili y in couple households (e.g. Plau 2006; an de Klis and Mulde
2008). In gene al, hey do no con as hei esul s wi h di e ences in high mobili y
be ween male and emale singles, hough possible di e ences among singles would indi-
ca e gende di e ences ha go beyond dynamics in ela ionships.
The e o e, he subsequen analysis examines he ollowing hypo hesis: high mobili y
di e s be ween women and men no only in ela ionships wi h and wi hou child en bu
also among singles, al hough hey do no ha e o coo dina e wi h o adap o a pa ne .
Thus, ou analysis b oadens he scope o in e p e a ion o gende di e ences in high mobil-
i y among people in ela ionships, since he di e ences be ween men and women hen ha e
o be in e p e ed as he sum o in luences dependen on and independen o ela ionships
and pa en hood. A he same ime, his means ha a deepe explana o y con ex o emale
and male decision-making p ocesses has o be used o expand he hypo heses o gende
oles, amily economics, nego ia ion p ocesses and pa ia chal powe ela ions. Addi ion-
ally, we in es iga e how indi idual and household- ela ed a ibu es in e ac wi h gende
1743
T anspo a ion (2022) 49:1737–1764
1 3
and how hese in e ac ions di e be ween singles and people in pa ne ships wi h and wi h-
ou child en.
Da a andme hodology
Da a
This esea ch uses he Ge man Family Panel pai am (“Panel Analysis o In ima e Rela-
ionship and Family Dynamics”) Release 10.0 (B üde l e al. 2019). This mul idisciplina y
annual long- e m s udy s a ed in 2008 and ocuses on pa ne ship and amily dynamics
among pe sons o he 1971–73, 1981–83 and 1991–93 bi h coho s. A special ea u e o
pai am is he mul i-ac o app oach: in addi ion o he andomly selec ed pe sons— he
‘ancho s’— hei pa ne s a e also in e iewed. This p o ides an ex ao dina y amoun o
in o ma ion abou he ancho and, whe e applicable, abou he pa ne (see Huinink e al.
2011 o a de ailed desc ip ion o he s udy). As he p esen esea ch examines gende di -
e ences in wo k- ela ed mobili y, i only uses su ey wa es 1, 3, 5 and 7 o 10, which
con ain in o ma ion on commu ing ime. Due o he esea ch ocus, he p esen s udy only
e e s o ancho s wi h he p ima y ac i i y s a us “ ull- ime-employed” o “pa - ime-
employed”. Gi en he ac ha sel -employed people ha e g ea e con ol o e hei choice
o wo kplace ( an Omme en and an de S aa en 2008; Reuschke and Hous on 2020),
gene ally ha e sho e commu es, and gende di e ences in commu ing ime a e signi -
ican ly smalle o no signi ican (e. g. Chidamba am and Scheine 2020; Reuschke and
Hous on 2020), we decided o exclude hem om he analyses. Un o una ely, he pai am
s udy does no include all he in o ma ion lis ed in Table1 o iden i y he pa ne s’ mobil-
i y beha io .3 Fo his eason, he ancho s’ pa ne s a e no included in he analyses as sep-
a a e cases. Thus, he analyses a e limi ed o employed singles and o employed ancho s in
pa ne ships ega dless o he pa ne ’s labo o ce s a us. Consequen ly, he s udy exam-
ines gende di e ences in high mobili y among ancho s ac oss di e en households a he
han gende di e ences wi hin pa ne ships.
Since he numbe o highly mobile employees in he single su ey wa es is oo small
o a longi udinal analysis, we conduc ed a c oss-sec ional analysis, pooling da a om
he se en su ey wa es including commu ing ime. Acco dingly, he da a may include up
o se en obse a ions o he same ancho pe son. The pooled sample includes a o al o
51,470 obse a ions o ancho s. 26,211 o hese obse a ions ela e o ancho s wi hou he
p ima y ac i i y s a us “ ull- ime-employed” o “pa - ime-employed” and a e he e o e
excluded. Based on he esea ch app oach explained in he ollowing sec ion, 943 obse -
a ions o employed singles wi h child en a e also excluded om he analyses as well as
124 obse a ions om ancho s wi h missing in o ma ion abou ela ionship s a us and chil-
d en in he household. The inal sample he e o e encompasses 24,192 (weigh ed 22,201)
obse a ions.
3 None o he su ey wa es p o ides in o ma ion whe he he pa ne ’s wo kplace changes o no . The
same applies o he equency and du a ion o commu ing om he second esidence. In con as , he o he
da a (nigh s spen away om home o wo k- ela ed easons, he equency and du a ion o commu ing om
he main esidence, and he numbe o nigh s spen a he empo a y esidence) we e su eyed a i egula
in e als and pa ly in di e en wa es.
1744
T anspo a ion (2022) 49:1737–1764
1 3
Me hodology
As "High mobili y and gende di e ences" and "Gende di e ences in high mobili y in he
con ex o ela ionship s a us and household" ou line, he mos widesp ead hypo heses and
exis ing esea ch e idence iden i y a s ong link be ween pa ne ship and pa en hood and gen-
de di e ences in high mobili y. This sugges s ha gende di e ences in high mobili y a ise
mainly (al hough no exclusi ely) om ans o ma ions in he pa ne ship and pa en hood
con ex . Ideally, a longi udinal app oach would be equi ed o in es iga e he complex p o-
cess o becoming highly mobile, which in ol es decisions ac oss se e al empo al scales and
in ol es di e en domains o li e, e.g. esiden ial and job loca ion. While ou da a is no sui ed
o conduc a longi udinal s udy ("Da a"), we po ay h ee impo an li e s ages wi h ega d o
gende di e ences in high mobili y by di iding he employees in o h ee g oups di e en ia ed
by pa ne ship and pa en hood s a us. The i s g oup consis s o employees wi h pa ne s and
child en (li ing in he same household) and he second g oup o employees wi h pa ne s bu
wi hou child en. These wo g oups also include ancho s who do no li e wi h hei pa ne s
in he same household, as we assume ha hese pa ne s also in luence he ancho s’ mobili y.4
The hi d g oup consis s o single employees wi hou child en. Fo each o hese g oups, we
conduc c oss-sec ional analyses and compa e he esul s be ween he g oups. This enables us
o d aw conclusions abou gende di e ences in high mobili y ha a e (in)dependen o pa -
ne ship and pa en hood. While gende oles, pa ia chal powe ela ions, amily and household
economics and he associa ed nego ia ion p ocesses a e expec ed o be pa icula ly impo an
o employees wi h pa ne s and child en, singles wi hou child en can decide mos ly inde-
penden ly and do no ha e o balance hei own in e es s agains hose o hei pa ne s and
child en (Hu 2020). Thus, gende di e ences among singles a e mo e likely o be associ-
a ed wi h in e nalized gende di e ences, occupa ional seg ega ion and economic dispa i ies,
which may also be shaped by pa ia chal s uc u es, han wi h he o he heo e ical app oaches.
They hus se e as a kind o benchma k. Un o una ely, due o small sample sizes, we ha e
o e ain om he sepa a e analysis o one- and wo-wo ke couples and we ha e o exclude
singles wi h child en om he analysis.
Bina y logis ic models o each g oup es ima e he p obabili y o high mobili y. Howe e ,
he non-independen (clus e ed) obse a ions esul ing om he pooling iola e a undamen-
al equi emen o “o dina y” eg ession models. Igno ing his dependency would esul in
e o s in in e en ial s a is ics (Hu e al. 1998). Thus, we use he Gene alized Es ima ing Equa-
ion Me hodology (GEE), which akes he in e dependencies o obse a ions om one pe -
son in o accoun using a wo k co ela ion ma ix o wi hin-subjec dependencies as pa o
he model. The p esen ed analyses a e pe o med wi h a obus a iance es ima ion algo i hm
and he wo king co ela ion ma ix ype "independen " (see Zo n 2001; Ghisle a and Spini
2004; Ag es i 2007 o a de ailed desc ip ion o he GEE me hodology). The in e p e a ion o
he GEE model p esen ed he e co esponds o ha o o dina y logis ic eg essions. Howe e ,
he e is no coe icien o de e mina ion o GEE models. SPSS p o ides he Quasi Likelihood
unde Independence Model C i e ion (QIC), which is an ex ension o he Akaike In o ma-
ion C i e ion (AIC) o epea ed measu emen s. The Co ec ed Quasi Likelihood unde Inde-
pendence Model C i e ion (QICC) penalizes small sample size and model complexi y. Models
wi h smalle QIC(C) a e be e . (Ga son 2013) We p o ide he QIC(C) o ou models and he
4 0.1% o ancho s wi h child en li e in a same-sex pa ne ship. The p opo ion o ancho s wi hou child en
who li e in a same-sex pa ne ship is subs an ially highe (3.3%). Analyses ha exclude hese ancho s do
no lead o no ewo hy changes in he esul s. Fo his eason, he analyses include also ancho in same sex-
pa ne ships.
1751
T anspo a ion (2022) 49:1737–1764
1 3
Employees wi h a leas one p eschool-age child a e less likely o be highly mobile
(males 18.6%, emales 6.3%) han employees wi h olde child en (males 22.5%, emales
8.3%). Employees wi h g ea e in ol emen in childca e a e less highly mobile, especially
women. Women wi h main childca e esponsibili ies a e conside ably less likely o be
highly mobile (4.8%) han women who sha e childca e wi h hei pa ne s in an equal way
(8.9%) and women whose pa ne s a e mainly esponsible o childca e (15.0%). Thus, gen-
de di e ences—measu ed by he a io o highly mobile men o women—become la ge
wi h highe in ol emen in childca e. Rega dless o gende , employees wi h pa ne s and
child en who a e sole o main wage ea ne s a e mo e o en highly mobile (males22.2%,
emales6.3%) han equal o seconda y wage ea ne s (males16.1%, emales6.1%). By con-
as , he e is a he e ogeneous pa e n among childless employees in pa ne ships. While
he e is no clea ela ion be ween income cons ella ion and high mobili y o men, emale
equal o second wage ea ne s wi hou child en a e mo e o en highly mobile (16.1%) han
emale single o main wage ea ne s wi hou child en (13.5%).
The p opo ion o highly mobile people ends o inc ease wi h income, especially among
women, so gende di e ences a e gene ally smalle among employees wi h highe incomes.
In addi ion, employees wi h high oca ional deg ees a e mo e o en highly mobile. The
same is ue o employees wi h empo a y con ac s excep o women wi h pa ne s and
child en (5.9% wi h s. 6.9% wi hou empo a y con ac ). Howe e , he p opo ion o em-
po a y employees wi h pa ne s and child en is e y low compa ed o he o he g oups.
Especially o emale singles, empo a y employmen seems o be associa ed wi h g ea e
p obabili ies o high mobili y (18.9% s. 9.8% wi hou empo a y con ac ). Male employ-
ees in he whi e-colla sec o a e mos likely o p ac ice high mobili y, ollowed by employ-
ees in he blue and g een-colla and pink-colla sec o s. The si ua ion o women is mo e
incohe en : among women wi h pa ne s and child en, employees in he whi e-colla sec o
a e mos equen ly highly mobile (24.7%). In con as , emale employees wi hou child en
in he whi e- and pink-colla sec o a e almos equally highly mobile. Consequen ly, he
p opo ion o highly mobiles among childless male and emale employees wo king in he
pink-colla sec o is nea ly balanced. In he ollowing sec ion, eg ession analysis aims o
disen angle hese co ela ions.
Mul i a ia e analysis
Table4 displays o dinal eg ession models o he h ee g oups o employees di e en i-
a ed by pa ne ship and pa en hood. As he li e a u e indica es one- ailed hypo heses o
mos o he independen a iables (Table2), Table4 con ains mos ly one- ailed p- alues.
Di ision o childca e and age o child( en) a e only ele an o employees wi h a pa ne
and child en and a e included in Model1. Fo he sake o compa abili y wi h childless
employees, Model2 does no conside hese a iables. Models1 and 2 o employees wi h
a pa ne and child en e eal la gely he expec ed co ela ion wi h simila magni ude wi h
one excep ion: conside a ion o he di ision o childca e in Model1 no iceably educes
he Exp(B) o gende .7 Howe e , e en aking childca e in o accoun , men a e signi ican ly
mo e o en highly mobile han women, all else being equal (Exp(B) > 1). Thus, while he
7 Excluding "di ision o childca e" in Model 1, he Exp(B) o gende is 0.215, and hus a abou he same
le el as in Model 2.

1752
T anspo a ion (2022) 49:1737–1764
1 3
Table 4 Logis ic eg ession analyses o high mobili y di e en ia ed by pa ne ship and pa en hood
Wi h pa ne and child Wi h pa ne w/o
child
Singles w/o
child
Wi h pa ne and child Wi h pa ne
w/o child
Singles w/o child
Model 1 Model 2 Model 3 Model 4 Model 1a Model 2a Model 3a Model 4a
Exp (B) Sig.aExp (B) Sig.aExp (B) Sig.aExp (B) Sig.aExp (B) Sig.aExp (B) Sig.aExp (B) Sig.aExp (B) Sig.a
In e cep 0.087 0.000 0.063 0.000 0.094 0.000 0.091 0.000 0.080 0.000 0.057 0.000 0.061 0.000 0.044 0.000
Male ( e .
Female) 1.692 0.003 2.120 0.000 1.412 0.003 1.644 0.007 2.724 0.039 3.178 0.004 3.613 0.000 5.193 0.000
A leas one child < 6yea s ( e . No)
Yes 0.802 0.060b- - - - - - 1.129 0.562b- - - - - -
Male * yes - - - - - - - - 0.656 0.089b- - - - - -
(Men: yes) - - - - - - - - (0.740) - - - - - - -
Di ision o childca e ( e . Mainly he pa ne )
Egali a ian 0.929 0.279 - - - - - - 0.611 0.117 - - - - - -
Mainly ancho 0.590 0.010 - - - - - - 0.378 0.012 - - - - - -
Male * egali a ian - - - - - - - - 1.562 0.303b- - -
Male * mainly
ancho
- - - - - - - - 1.494 0.529b- - - - - -
(Men: egali a ian) - - - - - - - - (0.955) - - - - - -
(Men: mainly
ancho )
- - - - - - - - (0.565) - - - - - -
Income cons ella ions ( e . sole o main wage ea ne )
Equal o second-
a y
0.856 0.225b0.852 0.185b1.146 0.195b- - 1.442 0.205b0.962 0.867b1.329 0.089b- -
Male * equal o
seconda y
- - - - - - - - 0.530 0.051b0.852 0.562b0.794 0.289b- -
(Men: equal o
seconda y)
- - - - - - - - (0.764) (0.820) (1.055) - -
1753
T anspo a ion (2022) 49:1737–1764
1 3
Table 4 (con inued)
Wi h pa ne and child Wi h pa ne w/o
child
Singles w/o
child
Wi h pa ne and child Wi h pa ne
w/o child
Singles w/o child
Model 1 Model 2 Model 3 Model 4 Model 1a Model 2a Model 3a Model 4a
Exp (B) Sig.aExp (B) Sig.aExp (B) Sig.aExp (B) Sig.aExp (B) Sig.aExp (B) Sig.aExp (B) Sig.aExp (B) Sig.a
Pe sonal mon hly ne income ( e . < 1,200 €)
€1,200–1,699 1.721 0.003 1.753 0.002 1.245 0.074 1.386 0.041 2.483 0.000 2.477 0.000 1.620 0.015 1.669 0.054
€1,700–2,299 2.480 0.000 2.650 0.000 2.225 0.000 1.950 0.002 3.708 0.000 3.891 0.000 3.444 0.000 2.398 0.020
> €2,300 3.127 0.000 3.367 0.000 3.134 0.000 2.197 0.004 2.856 0.003 2.968 0.001 4.038 0.000 3.838 0.004
Male * €1,200–
1,699
- - - - - - - - 0.376 0.004 0.369 0.003 0.517 0.015 0.789 0.272
Male * €1,700–
2,299
- - - - - - - - 0.404 0.011 0.403 0.011 0.376 0.002 0.742 0.277
Male * > €2,300 - - - - - - - - 0.732 0.253 0.725 0.238 0.514 0.044 0.437 0.088
(Men: €1,200–
1,699)
- - - - - - - - (0.933) (0.914) (0.838) (1.317)
(Men: €1,700–
2,299)
- - - - - - - - (1.499) (1.568) (1.293) (1.780)
(Men: > €2,300) - - - - - - - - (2.091) (2.153) (2.076) (1.676)
Highes oca ional deg ee ( e . < uni e si y)
> Uni e si y 0.775 0.067 0.723 0.022c0.883 0.229 0.933 0.401 0.497 0.025c0.481 0.015c0.796 0.181 1.289 0.274
Male * ≥ uni e -
si y
- - - - - - - - 1.756 0.165 1.799 0.066 1.193 0.302 0.534 0.117
(Men: ≥ uni e -
si y)
- - - - - - - - (0.873) (0.865) (0.949) (0.688)
Tempo a y con ac ( e . No)
Yes 1.319 0.056 1.292 0.064 1.423 0.003 1.770 0.000 0.712 0.114 0.802 0.189 1.481 0.015 2.570 0.003
Male * yes - - - - - - - - 2.509 0.009b2.110 0.023b0.982 0.944b0.586 0.170b
1754
T anspo a ion (2022) 49:1737–1764
1 3
Table 4 (con inued)
Wi h pa ne and child Wi h pa ne w/o
child
Singles w/o
child
Wi h pa ne and child Wi h pa ne
w/o child
Singles w/o child
Model 1 Model 2 Model 3 Model 4 Model 1a Model 2a Model 3a Model 4a
Exp (B) Sig.aExp (B) Sig.aExp (B) Sig.aExp (B) Sig.aExp (B) Sig.aExp (B) Sig.aExp (B) Sig.aExp (B) Sig.a
(Men: yes) - - - - - - - - (1.786) (1.693) (1.455) (1.505)
Employmen sec o ( e . Whi e-colla )
Blue and g een-
colla
0.754 0.038 0.719 0.023 0.854 0.159 0.749 0.095 0.363 0.003 0.361 0.002 0.509 0.040 1.171 0.373
Pink-colla 0.581 0.001 0.560 0.001 0.896 0.243 0.715 0.086 0.571 0.024 0.553 0.013 1.133 0.282 1.616 0.111
Male * blue and
g een-colla
- - - - - - - - 2.406 0.039b2.460 0.019b1.672 0.225b0.472 0.162b
Male * pink-
colla
- - - - - - - - 1.023 0.949b1.100 0.785b0.553 0.057b0.244 0.003b
(Men: blue and
g een-colla )
- - - - - - - - (0.849) (0.888) (0.851) (0.553)
(Men: pink-colla ) - - - - - - - - (0.584) (0.608) (0.627) (0.394)
QIC, in e cep 8,085 8,085 7,342 4,066 8,085 8,085 7,342 4,066
QICC, in e cep 8,053 8,053 7,337 4,060 8,053 8,053 7,337 4,060
QIC 6,109 6,413 6,439 3,627 6,095 6,399 6,437 3,620
QICC 6,068 6,379 6,401 3,585 6,035 6,350 6,375 3,555
R2 (Cox & Snell)d6.9 6.6 3.0 2.5 7.6 7.0 3.6 3.4
R2 (Nagelke ke)d12.2 11.7 4.9 4.3 13.4 12.6 5.9 5.9
n (obse a ions) 10,061 10,789 6,601 4,073 10,601 10,789 6,601 4,073
n (indi iduals) 3,426 3,598 3,140 2,000 3,426 3,598 3,140 2,000
a one- ailed p- alues b wo- ailed p- alues cno signi ican due o he one- ailed p- alues/hypo hesis o a posi i e associa ion be ween a high oca ional deg ee and high mobili y
dR2 o o dina y logis ic eg ession
1755
T anspo a ion (2022) 49:1737–1764
1 3
sha e o childca e is nega i ely ela ed o high mobili y, i only con ibu es o a limi ed
ex en o gende di e ences.
In line wi h he esul s o Reuschke and Hous on (2020), Model 1 shows no signi ican
associa ion be ween he age o child en and high mobili y among pa en s. As assumed,
highe income and employmen in he whi e-colla sec o is posi i ely associa ed wi h high
mobili y (Model 1 and 2). In con as , he models do no con i m he expec ed posi i e co -
ela ion be ween high mobili y and a high oca ional deg ee while aking income cons el-
la ion, income and he employmen sec o in o accoun .8 Fu he mo e, employees wi h a
pa ne and child en on empo a y con ac s a e no signi ican ly mo e likely o be highly
mobile, which could be due o he low p opo ion on empo a y con ac s ("Desc ip i e
analysis"). Addi ionally, he income cons ella ions in pa ne ships do no ha e any signi i-
can co ela ion wi h high mobili y o employees wi h pa ne s and child en.
Model3 o childless wo ke s in pa ne ships and Model4 o singles show simila bu
weake co ela ions ega ding gende and income. The same is ue o employmen sec o ,
especially he pink-colla sec o , bu he associa ions a e no signi ican o he wo g oups
wi hou child en. In con as o Models1 and2 o employees wi h child en, he posi i e
associa ion be ween high mobili y and empo a y con ac s is signi ican o childless
employees. Simila ly o employees wi h pa ne s and child en, he e is no signi ican co -
ela ion be ween a high oca ional deg ee and he income cons ella ion wi h high mobili y
in he o he g oups.
In gene al, he models ha include only main e ec s con i m mos o he expec ed co -
ela ions. Rega dless o g oup classi ica ion, he known gende di e ences in high mobil-
i y a e signi ican , al hough hey a e highes among employees wi h pa ne s and child en
(Model2). Howe e , when con olling o childca e (Model1), he gende di e ences in
his g oup a e simila o hose obse ed among childless employees. Mo eo e , i is s ik-
ing ha single women who ha e li le o no need o consul a ion o coo dina ion wi h a
pa ne a e also less likely o be highly mobile han men (Model4). These gende di e -
ences e en seem o be sligh ly g ea e han o childless wo ke s wi h pa ne s (Model3).
The in e ac ion e ms in Models1a–4a p o ide u he indings o in e p e a ion. Due
o he gende in e ac ion, he condi ional main e ec s indica e he Exp(B) only o women
and he espec i e gende in e ac ion e ms he Exp(B) o men ela i e o women. Acco d-
ingly, hei p oduc equals he Exp(B) o men, which a e shown in b acke s in Table4.
Fu he mo e, due o he in e ac ions wi h gende , he main condi ional e ec o gende
e e s only o pe sons wi h he cha ac e is ics o all e e ence ca ego ies and hus o a e y
speci ic g oup o pe sons. Consequen ly, he condi ional main e ec o gende is only o
limi ed ele ance. Due o he la ge numbe o possible combina ions o cha ac e is ics con-
side ed he e, we e ain om p esen ing he gende di e ences o each o hem and con-
cen a e on whe he he gende in e ac ion e ms wi h he espec i e cha ac e is ic inc ease
(Exp(B) > 1.0) o dec ease (Exp(B) < 1.0) gende di e ences in high mobili y. As discussed
in "Me hodology", we ocus p ima ily on he e ec s and less on he signi icances.
To exempli y he in e p e a ion, conside Model3a o childless employees wi h pa -
ne s. The odds o high mobili y o emale employees in he highes income ca ego y
(> €2,300) a e 4.038 imes highe han he odds o emale employees in he lowes income
ca ego y ( e e ence ca ego y < €1,200). The Exp(B) o he in e ac ion e m male* > €2,300
is 0.514 and conside ably weakens he condi ional main e ec o men. Consequen ly, he
8 Reg essions wi hou hese a iables, no epo ed he e o he sake o b e i y, con i m a posi i e signi i-
can co ela ion be ween a high oca ional deg ee and high mobili y.
1756
T anspo a ion (2022) 49:1737–1764
1 3
Exp(B) o male employees in he highes income ca ego y a e 4.038*0.514 = 2.076. Thus,
especially o women, a high income has a posi i e co ela ion wi h high mobili y, which
also a ec s gende di e ences: o employees wi h cha ac e is ics o all e e ence ca ego-
ies, including an income below €1,200, he condi ional Exp(B) o gende is 3.613. Keep-
ing all else unchanged excep he income, he Exp(B) o gende o employees in he high-
es income g oup is 3.613*0.514 = 1.857 and hence subs an ially lowe .
Models1a and2a o employees wi h pa ne s and child en including gende in e ac-
ion e ms again show simila co ela ions. The majo i y o he condi ional main e ec s a e
s onge o women han in he main-e ec Models1 and2 (Exp(B) de ia e u he om 1)
and a e weakened o men by gende in e ac ions (Exp(B) a e close o 1). The condi ional
e ec o gende dec eases when childca e is aken in o accoun . Howe e , as men ioned
abo e, his e ec is linked o many cha ac e is ics and is he e o e o limi ed in o ma i e
alue.
Model1a indica es a sligh ly posi i e ough no signi ican associa ion be ween he p es-
ence o child en younge han 6yea s old and high mobili y o women. Fo men, his asso-
cia ion seems o be a he nega i e. Fu he mo e, Model1a shows ha e en i women and
men con ibu e he same amoun o childca e, childca e is o a g ea e ex en nega i ely
associa ed wi h women’s high mobili y and hus inc eases gende di e ences. Egali a ian
childca e seems o be nega i ely ela ed o high mobili y only o women. I he ancho is
mainly esponsible o childca e, his is associa ed wi h lowe p obabili ies o high mobil-
i y, especially o women. By con olling childca e in Model1a, gende di e ences end
o be smalle o equal and seconda y wage ea ne s, as he s a us o equal and seconda y
wage ea ne s ends o be posi i ely associa ed wi h high mobili y o women, while he
opposi e seems o be he case o men. Models1a and 2a indica e ha high income is
associa ed wi h high mobili y, pa icula ly o women. Thus, gende di e ences in highe
income g oups a e smalle han in he lowes . The models also indica e ha he posi i e
associa ion o empo a y con ac s wi h high mobili y in he main-e ec Models1 and2 is
due o he associa ion o men. Only men wi h a empo a y con ac seem o be mo e likely
o be highly mobile, bu no women. On he o he hand, he nega i e co ela ion be ween
employmen in he blue and g een-colla sec o and high mobili y ha was iden i ied in he
main-e ec Models1 and2 is p ima ily p esen among women.
The espec i e gende in e ac ions a y be ween he g oups o employees subdi ided
by pa ne ship and pa en hood s a us, whe eby all in e ac ion e ms o singles a e associ-
a ed wi h lowe gende di e ences (Exp(B) < 1). The compa ison o he models includ-
ing gende in e ac ions e eals di e en pa e ns. The i s pa e n is isible o pe sonal
ne income. Rega dless o g oup, income is mo e posi i ely associa ed wi h high mobil-
i y o women, whe eby gende in e ac ion ends o decline wi h childlessness and single
s a us (wi h he excep ion o he highes income class). The declining in e ac ion e ec s
a e mainly due o he declining posi i e co ela ion o income and high mobili y among
women, while he Exp(B) o income among men di e less ac oss he g oups o employ-
ees. This pa e n also occu s o he income cons ella ion when childca e is conside ed.
He e, oo, he in e ac ion o gende and s a us as equal o second ea ne is sligh ly mo e
p onounced among employees wi h child en (Model1b) han among employees wi hou
child en (Model3b) and is hus mo e s ongly co ela ed wi h a dec ease in gende di e -
ences among he o me .
A e e se pa e n occu s in he highes income ca ego y. He e he in e ac ion wi h gen-
de inc eases wi h childlessness and single s a us. Howe e , he numbe o emale employ-
ees wi hin he highes income class is limi ed (Table3). The hi d pa e n conce ns he
highes le el o oca ional quali ica ions, empo a y con ac s and employmen sec o .

1757
T anspo a ion (2022) 49:1737–1764
1 3
While he espec i e gende in e ac ion e ms a e posi i e (Exp(B) > 1) o employees wi h
pa ne s and child en, who a e associa ed wi h highe gende di e ences compa ed o he
espec i e e e ence g oups, he in ensi y o he posi i e in e ac ions e ec s dec eases and
becomes he opposi e wi h childlessness and single s a us (Exp(B) < 1). Once again, his is
p ima ily due o he changed associa ion o emale employees in he g oups o employees.
Female singles wi h empo a y con ac s a e signi ican ly mo e mobile han emale singles
wi h pe manen con ac s. Fo men he co ela ion seems o be much weake . Mo eo e ,
he lowe p obabili y o high mobili y among employed singles in he g een and blue as
well as in he pink-colla sec o compa ed o whi e-colla sec o , which was ound in he
main-e ec Model4, is hus limi ed o single men. Acco dingly, gende di e ences in high
mobili y among single employees wi h a high oca ional deg ee, empo a y con ac and
employmen in he blue and g een-colla and in he pink-colla sec o a e lowe han in he
espec i e e e ence ca ego ies.
Measu ed by R2 o con en ional eg essions, he model i is highes o he models o
he employees wi h a pa ne and child en. A 12% hey a e low, bu ypical o s udies o
anspo beha io a he indi idual le el (Chidamba am and Scheine 2020). Fo childless
employees wi h and wi hou pa ne s, he R2 alue is app oxima ely as low. Fo all g oups,
he inclusion o in e ac ion e ms sligh ly inc eases he model i .
Discussion andconclusions
This esea ch examines gende di e ences in high mobili y. To b oaden he scope o in e -
p e a ion o gende di e ences in high mobili y in couples, we also include single employ-
ees in he analyses. Fu he mo e, using gende in e ac ion e ms as pa o eg ession
models allows us o in e how di ision o labo and socioeconomic and occupa ional cha -
ac e is ics in e ac wi h gende in e ms o high mobili y, conside ing di e en li e s ages.
Thus, ou esea ch combined wi h exis ing and u u e esea ch can p o ide he basis o
mo e p ecise a ge -g oup speci ic policy implica ions.
When in e p e ing he esul s, a en ion should be paid o he explana ions and hypo h-
eses o gende di e ences in high mobili y, bu he assessmen o high mobili y ("High
mobili y and gende di e ences") also has o be conside ed. High mobili y p ac ices can
be a s a egy o a ca ee s a o p o essional ad ancemen in ce ain p o essions and could
acili a e access o (mo e a ac i e) jobs. In his sense, high mobili y is an ad an age and
he lowe a e o high mobili y among women may indica e es ic ions in hei p o es-
sional ca ee s. Howe e , whe he high mobili y is an ad an age also depends on he ype o
ca ee . Fu u e esea ch wi h a mo e di e en ia ed subdi ision o occupa ions han occupa-
ional sec o s could p o ide u he insigh s. Ne e heless, high mobili y is also associa ed
wi h social and heal h bu dens. A possible al e na i e o high mobili y is eloca ion, which
may in u n lead o a g ea e need o high mobili y o he pa ne o o up oo ing he am-
ily. The e o e, he decision o highly mobile p ac ices ins ead o eloca ion elie es no
only he pa ne bu also he child en, who can s ay in he desi ed o amilia esiden ial and
wo king en i onmen . In his sense, high mobili y is a way o combining ca ee and pa -
ne ship o amily, bu i is also a bu den o hose who p ac ice i . The lowe le el o high
mobili y among women should he e o e no be in e p e ed exclusi ely as a disad an age.
The esul s p esen ed do no p o ide de ini i e e idence on how gende di e ences
in high mobili y should be e alua ed. Howe e , hey pa ially suppo he pe cep ion o
high mobili y as a bu den. Acco ding o he desc ip i e esul s, singles who a e la gely
1758
T anspo a ion (2022) 49:1737–1764
1 3
independen in hei decisions a e less likely o be highly mobile han childless employ-
ees wi h pa ne s. This applies o bo h men and women, indica ing ha hey a oid high
mobili y i possible. A ela ionship gene a es s onge spa ial ies, which employees may
compensa e o wi h high mobili y. The e o e, and acco ding o he desc ip i e esul s, bo h
gende s a e mo e likely o be highly mobile in pa ne ships wi hou child en han singles,
al hough his is sligh ly mo e he case o women. Consequen ly, gende di e ences a e
ma ginally smalle among childless employees wi h pa ne s han among singles. The mul-
i a ia e analyses also poin o hese smalle gende di e ences. Child en in he household
inc ease he local a achmen e en mo e. The desc ip i e as well as he eg ession models
indica e ha he p opo ion o employees wi h highly mobile p ac ices emains abou he
same among men, whe eas he p opo ion o women is conside ably lowe . As a esul ,
gende di e ences a e g ea es among employees wi h pa ne s and child en. Taking child-
ca e in o accoun educes hese gende di e ences and hus suppo s HRH and he demand
o policy ac ion o suppo he econcilia ion o wo k and amily li e. To acili a e his ec-
oncilia ion and a oid high mobili y, policy make s and companies could o e in addi ion o
comp ehensi e childca e se ices, mo e lexible wo king condi ions o employees. These
include wo king om home, lexible wo king hou s and he (pa ial) eplacemen o ace-
o- ace mee ings and he associa ed business ips by ideo con e ences. As he COVID-19
c isis has shown, his is possible o a high deg ee in many, hough no all, occupa ions.
Howe e , i is wo h no ing ha he di ec ion o causali y is no clea . On he one hand,
wo ke s migh a oid high mobili y o ca e o child en; on he o he hand, high engage-
men in childca e migh coun e ac high mobili y. Fu u e s udies could in es iga e causal-
i y using s uc u al equa ion modeling. Ne e heless, signi ican gende di e ences emain
when childca e is conside ed, which could be a ibu ed o pa ia chal powe ela ions and
he gende ed di ision o labo besides childca e in couples. Howe e , as hese emaining
gende di e ences ha e app oxima ely he same magni ude as o singles, i is easonable
o assume ha hese gende di e ences a e due o gende -speci ic ac o s ha a e al eady
a wo k o singles. These di e ences may be based on occupa ional seg ega ion and eco-
nomic dispa i ies, o o a gene al lowe accep ance o women owa ds high mobile p ac-
ices (in e nalized gende p e e ences). I he la e we e he case, women would ejec high
mobili y mo e han men and he lowe le el o high mobili y o women in pa ne ships
would cons i u e an ad an age o women.
The eg ession models including gende in e ac ion e ms p o ide supplemen a y ind-
ings o in e p e a ion. Al hough he esul s a e a he explo a o y in na u e ("Me hod-
ology") and he desc ibed di e ences be ween he employmen g oups di e en ia ed by
pa ne ship and pa en hood s ill need o be es ed o signi icance (e.g. by using h ee-way-
in e ac ions, "Va iables"), he esul s poin o gende -speci ic di e ences ha in ensi y o
diminish wi h inc easing bonding h ough a pa ne ship and pa en hood. These changes
a ise p ima ily om al e a ions in he associa ions o women, sugges ing ha pa ne ship
and pa en hood a e pa icula ly associa ed wi h emale a el beha io .
The gende in e ac ion wi h empo a y con ac s is in e es ing o he e alua ion o high
mobili y as a (dis)ad an age. Among singles, empo a y con ac s ha e a much highe pos-
i i e co ela ion wi h high mobili y o women han o men. Con e sely, his means ha
women in pe manen employmen a oid high mobili y mo e han men and mainly p ac-
ice high mobili y unde he condi ions o empo a y con ac s. Fo emale employees wi h
pa ne s, he co ela ion be ween empo a y con ac s and high mobili y is much less p o-
nounced and a abou he same le el as o men. I seems ha empo a y con ac s a e a
less decisi e ac o o high mobili y han he compa ibili y o a ela ionship and ( wo)
p o essional ca ee (s). By con as , among employees wi h pa ne s and child en, a posi i e
1759
T anspo a ion (2022) 49:1737–1764
1 3
co ela ion be ween empo a y con ac s and high mobili y exis s only o men. This could
be due o and consis en wi h he gende ole o men as “b ead-winne s” in ha men (usu-
ally he main ea ne s) who ha e empo a y con ac s and a e in pa ne ships wi h child en
p e e o p ac ice high mobili y a he han o up oo he amily om i s amilia en i on-
men . Howe e , he p opo ion o empo a y employees among employees wi h child en
is conside ably lowe , implying ha employees do no choose o ha e child en un il hey
ha e “mo e secu e” (pe manen ) employmen a angemen s. Conside ing high mobili y as
a bu den and as a s a egy o b idge unce ain si ua ions, bo h he policy make s and he
companies could aim o educe empo a y employmen whe e possible.
Gende -speci ic economic dispa i ies could be one eason o he s onge posi i e co -
ela ion o income wi h high mobili y o women, e en among singles. Thus, a be e -paid
job o e s an incen i e o high mobili y, especially o women. In his sense, educing o ,
ideally, o e coming gende economic dispa i ies could educe incen i es o high mobil-
i y o women and he associa ed bu dens. The pa ne ship-independen di e ences in en-
si y wi h inc easing a achmen o pa ne s and child en p ima ily due o he inc easing
associa ions o women’s income wi h high mobili y. This migh poin o he impo ance o
women’s income wi hin he amily and household decisions and in he associa ed nego ia-
ion p ocesses. The gende in e ac ion wi h he income cons ella ion also ends o inc ease
wi h child en (when childca e is con olled o ) and implies ha emale equal and second-
a y wage ea ne s a e mo e o en highly mobile han hei male coun e pa s. I is likely ha
he household income o he income o he couple is highe han ha o sole o main wage
ea ne s. In his sense, he esul s a e consis en wi h Chidamba am and Scheine (2020),
who ound ha highe household income is associa ed wi h longe commu ing imes o
women and hus wi h lowe gende di e ences. They a ibu e his on he one hand o mo e
p onounced equal in a-couple economic powe ela ions, and on he o he hand o a pos-
sible sel -selec ion e ec , as mo he s who wo k migh be mo e ca ee -o ien ed, which in
u n is ela ed o longe commu es. He e, a di e en ia ion be ween equal and second ea n-
e s would ce ainly be in o ma i e, bu is no possible due o he small sample size ("Va i-
ables"). Howe e , he ac ha he s a us o equal o seconda y ea ne s is associa ed wi h
mo e high mobili y o women could also be due o main enance and child-suppo ing ips
on he way o wo k. These ips a e mo e equen o women han o men and canno be
dis inguished om "pu e" commu es o wo k based on pai am ("Va iables"). This, in addi-
ion o he sel -selec ion e ec , migh also explain why he p esence o p eschool child en
ends o be posi i ely ela ed o high mobili y o women, bu no o men.
The inding ha wo king in he pink-colla sec o ends o be posi i ely associa ed wi h
high mobili y o emale singles bu no o male singles weakens he hypo hesis o occu-
pa ional seg ega ion. Howe e , he gende in e ac ion wi h he pink-colla sec o changes
wi h inc easing a achmen o pa ne s and child en. Acco dingly, he e is no no ewo hy
in e ac ion be ween gende and he pink-colla sec o o employees wi h pa ne s and chil-
d en. Again, p ima ily o women, he e ec s a y di e en ia ed by pa ne ship and pa en -
hood. This indica es ha i is no occupa ional seg ega ion pe se ha is linked o gende
di e ences in high mobili y, bu a he he compa ibili y o employmen in he pink-colla
sec o and high mobili y wi h pa ne ship and pa en hood. Fo example, speci ic wo k-
ing condi ions such as la e o changing wo king hou s can coun e ac his compa ibili y.
The policy, bu also he companies hemsel es, should hus p omo e he compa ibili y o
amily and ca ee , especially in he pink-colla sec o . Howe e , ela i e cha ac e is ics in
pa ne ships may also explain why wo king in he pink-colla sec o is associa ed wi h a
lowe likelihood o high mobili y o mo he s. A mo e de ailed examina ion o hese ela-
i e measu es in couples ha a e decisi e o amily and household decisions could p o ide
1760
T anspo a ion (2022) 49:1737–1764
1 3
u he insigh s. We ha e e ained om his in a o o compa abili y wi h singles, bu
pai am o e s u he possibili ies o in es iga e his in mo e de ail.
The esul s, di e en ia ed by gende as well as by pa ne ship and pa en hood, indi-
ca e ha gende di e ences in high mobili y a e ela ed o pa ne ship and pa en hood as
well as o o he ac o s. Gende di e ences in high mobili y a e hus no only a esul o
nego ia ion p ocesses, o a (pa ia chal) powe s uc u e in ela ionships o o gende ed
labo di ision. They a e also an exp ession o in e nalized gende p e e ences, gende ed
occupa ional seg ega ion and economic dispa i ies independen o pa ne ships and pa -
en hood. Fu he quali a i e and quan i a i e esea ch is needed o disen angle hese
in luences. Fo he la e , he pai am s udy o e s he oppo uni y o examine a i udes,
sa is ac ion and p e e ences conce ning ime use and employmen si ua ion as well as
pa ne ship and amily. Addi ionally, indi idual-le el panel models would be desi able
o con i m and ex end he esul s p esen ed. These panel models could con ol o unob-
se ed he e ogenei y and associa ed endogenous selec ion e ec s, such as ca ee o i-
en a ion and p e e ences ega ding place o esidence and high mobili y. I would also
be concei able o di e en ia e be ween LDC, shu les and a i-mobiles, as hese di e
om one ano he despi e bo h being a way o o e come di icul ies in econciling p o-
essional and p i a e li e. Mo eo e , i would be in e es ing o compa e he co ela ions
iden i ied he e o Ge many wi h o he coun ies, e.g., Cen al and Eas e n Eu opean
coun ies, whe e he gende pay gap is less p onounced and he sha e o women in ull-
ime wo k is highe (Eu opean Ins i u e o Gende Equali y 2014; Boll and Lagemann
2018).
Acknowledgemen s This pape uses da a om he Ge man Family Panel pai am, coo dina ed by Jose
B üde l, Sonja D obnič, Ka s en Hank, F anz Neye , and Sabine Walpe . pai am is unded as long- e m
p ojec by he Ge man Resea ch Founda ion (DFG).
Au ho s’ con ibu ions Isabelle Wach e : Concep ualiza ion, Me hodology, Fo mal analysis and in es iga-
ion, W i ing; Ch is ian Holz-Rau: Supe ision.
Funding Open Access unding enabled and o ganized by P ojek DEAL.
Da a a ailabili y This pape uses da a om he Ge man Family Panel pai am, elease 10.0.0, which is
accessible o he scien i ic communi y as scien i ic use ile o schola ly analyses (h ps:// doi. o g/ 10. 4232/
pai am. 5678. 10.0.0).
Decla a ions
Con lic o in e es The au ho s decla e ha he e is no con lic o in e es .
Open Access This a icle is licensed unde a C ea i e Commons A ibu ion 4.0 In e na ional License,
which pe mi s use, sha ing, adap a ion, dis ibu ion and ep oduc ion in any medium o o ma , as long
as you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Com-
mons licence, and indica e i changes we e made. The images o o he hi d pa y ma e ial in his a icle
a e included in he a icle’s C ea i e Commons licence, unless indica ed o he wise in a c edi line o he
ma e ial. I ma e ial is no included in he a icle’s C ea i e Commons licence and you in ended use is no
pe mi ed by s a u o y egula ion o exceeds he pe mi ed use, you will need o ob ain pe mission di ec ly
om he copy igh holde . To iew a copy o his licence, isi h p:// c ea i eco mmons. o g/ licen ses/ by/4. 0/.