Ach , Ma in; Rebaudo, Ma a
A icle — Published Ve sion
A gende gap in gende gaps: social no ms and
housewo k epo ing
Empi ical Economics
P o ided in Coope a ion wi h:
Sp inge Na u e
Sugges ed Ci a ion: Ach , Ma in; Rebaudo, Ma a (2025) : A gende gap in gende gaps: social no ms
and housewo k epo ing, Empi ical Economics, ISSN 1435-8921, Sp inge , Be lin, Heidelbe g, Vol.
68, Iss. 6, pp. 2977-3029,
h ps://doi.o g/10.1007/s00181-024-02710-z
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/323197
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 p://c ea i ecommons.o g/licenses/by/4.0/
Empi ical Economics (2025) 68:2977–3029
h ps://doi.o g/10.1007/s00181-024-02710-z
A gende gap in gende gaps: social no ms and housewo k
epo ing
Ma in Ach 1,3 ·Ma a Rebaudo1,2
Recei ed: 16 May 2024 / Accep ed: 27 Decembe 2024 / Published online: 6 Feb ua y 2025
© The Au ho (s) 2025
Abs ac
Gende di e ences in he amoun o housewo k pe o med and he ole o social
no ms in explaining hese pe sis en gaps ha e ecei ed inc easing a en ion om
bo h policymake s and esea che s in ecen yea s. Howe e , no ms may no only
a ec he ac ual di ision o housewo k bu also po en ially in luence he epo ing
beha io in su eys. We s udy how e ospec i e esponses abou ime-use in ace-
o- ace in e iews a e in luenced by he gende o he in e iewe . Ou indings show
ha women end o epo signi ican ly mo e hou s o housewo k when in e iewed
by a woman a he han by a man. This e ec is no obse able o male esponden s,
esul ing in an in e iewe gende gap in he housewo k gende gap. Explo ing he
e ec in ela ion o se e al no m- ela ed cha ac e is ics indica es ha social no ms
play an impo an ole in he epo ing o housewo k hou s. The e o e, gende gap
es ima es based on ace- o- ace in e iews should be in e p e ed wi h g ea cau ion.
Keywo ds Gende ·Gende gap ·Doing gende ·Social no ms ·In e iewe e ec s
JEL classi ica ion D13 ·J16 ·D90
1 In oduc ion
The ole o social no ms as impo an d i e s o human beha io has a ac ed he
a en ion o esea che s no leas since he seminal wo k by Ake lo and K an on
(2000). In his ield o esea ch, a g owing li e a u e seeks o unde s and how social
The au ho s wan o hank semina pa icipan s a he EEA Cong ess, a he Annual Con e ence o he
Ve ein ü Socialpoli ik, a he RGS Doc o al Con e ence, Ma ianna Schaube and semina pa icipan s in
Sank Augus in, F eibu g and Gö ingen o help ul commen s and sugges ions.
BMa a Rebaudo
[email p o ec ed].de
1F aunho e Ins i u e o Applied In o ma ion Technology FIT, Sank Augus in, Ge many
2Uni e si y o F eibu g, F eibu g im B eisgau, Ge many
3Fede al Minis y o Finance, Be lin, Ge many
123
2978 M.Ach ,M.Rebaudo
no ms con ibu e o explaining pe sis en gende gaps in ea nings, ime-use and o he
ou comes. In hei p ominen con ibu ion, Be and e al. (2015) demons a e how
couples engage in no m-con o ming beha io by adjus ing hei labo ma ke ou comes
and ime spen on housewo k o adhe e o he adi ional male b eadwinne no m. In
his s udy, we o e a new pe spec i e on hese indings and analyze whe he and how
social no ms, which we de ine he e as socially p esc ibed gende no ms, a ec he
epo ing o housewo k hou s in su eys. Ou app oach exploi s a con ex ual ac o
po en ially ela ed o he ole o social no ms in he esponse p ocess: he gende o he
in e iewe . Speci ically, we es ima e esponden - ixed-e ec s models o s udy how
he epo ing o housewo k hou s changes when he in e iewe is emale as opposed
o male.
When s udying ime spen on housewo k, mos e alua ions ely on e ospec i e
ques ionnai e esponses a he han on dia y-based da a. The e emains, howe e ,
discussion abou how eliable such e ospec i e ime-use da a a e. Fo example,
Kan (2008) inds sys ema ic di e ences be ween ques ionnai e and dia y da a on
housewo k hou s which a y by gende and o he cha ac e is ics o he esponden s.
Se e al mechanisms o explain he gap be ween ques ionnai e-based and dia y-based
es ima es ega ding housewo k hou s ha e been p oposed in he li e a u e. On he
one hand, ques ionnai e-based es ima es a e likely o be biased due o an unclea
de ini ion o housewo k, ecalling p oblems o double-coun ing when housewo k is
pe o med simul aneously wi h o he ac i i ies (Kan 2008; P ess and Townsley 1998;
Wa ne 1986). On he o he hand, besides hese mo e echnical explana ions, epo ing
housewo k hou s in su eys is some imes conside ed a gende ed p ocess. P ess and
Townsley (1998) a gue ha social desi abili y plays a cen al ole in explaining di -
e ences be ween ques ionnai e and dia y es ima es, because some esponden s may
eel he p essu e o epo a le el o housewo k ha complies wi h no ma i e gende
oles. Ou pape ocuses on his aspec o housewo k epo ing and is, o he bes o
ou knowledge, he i s o p o ide empi ical e idence o whe he and how no ms
a ec he epo ing o housewo k hou s.
A g owing body o li e a u e poin s o he impo ance o social no ms in explaining
pe sis en gende di e ences in ime spen on housewo k ha canno be a ionalized
wi hin s anda d economic models. This ela es o indings abou how couples eac o
labo ma ke e en s (Fos e and S a on 2018), o child bi h o pa en al lea e (Schobe
and Zoch 2019; Schobe 2013), o en e ing e i emen (Leopold and Skopek 2015), and
also o he ole o ins i u ions in a ec ing gende di e ences (Cooke 2007; Lippmann
e al. 2020). One o he mos s iking o hese obse a ions has been demons a ed
by Be and e al. (2015): S anda d economic models o decision-making wi hin he
amily would p edic ha as a spouse’s con ibu ion o he couple’s income inc eases,
she should dec ease he housewo k hou s. Howe e , Be and e al. (2015) show ha
women who ou ea n hei pa ne s and he eby iola e he male b eadwinne no m,
ac ually inc ease hei ime spen on housewo k. This phenomenon has been in e -
p e ed in ligh o he “doing gende ” heo y i s p oposed by Wes and Zimme man
(1987). This heo y posi s ha indi iduals “do gende ” by ac i ely con o ming o soci-
e al expec a ions and no ms associa ed wi h hei pe cei ed gende iden i y. When a
adi ional gende ole is iola ed, o ins ance, when a wi e ou ea ns he husband,
123
A gende gap in gende gaps… 2979
indi iduals aim o compensa e o his iola ion by ein o cing hei gende iden i y
in o he ways, such as inc easing hei housewo k hou s.
The p esence o “doing gende ” beha io has been es ablished o se e al coun-
ies and con ex s (e.g., Be and e al. 2015; Lippmann e al. 2020; Bi man e al.
2003; Fos e and S a on 2018). The inding ha couples engage in no m-con o ming
beha io is no only ele an o he gende ed di ision o housewo k hou s, bu also
ex ends o women’s labo o ce pa icipa ion and income (Be and e al. 2015). How-
e e , ecen s udies by Ro h and Slo winski (2020) and Mu ay-Close and Heggeness
(2019) sugges ha he obse ed no m-con o ming beha io in his domain is he esul
o sys ema ic mis epo ing o incomes. In ligh o his, while “doing gende ” beha io
ega ding housewo k hou s may indeed be p esen , i could be in e ms o o e e-
po ing o housewo k hou s a he han ac ual beha io (Sulli an 2011). Then, he
a o emen ioned s a egies o adhe e o social no ms may s ill be ele an , bu p obably
in a di e en way han he li e a u e cu en ly sugges s.
Analogous o he wo k by Ro h and Slo winski (2020), who compa e o icial and
epo ed income da a o he same indi iduals, i would be ideal o compa e e ospec-
i e esponses abou ime-use o he “ ue” alues o see i ce ain cha ac e is ics lead
indi iduals o o e - o unde epo housewo k hou s. As such da a a e no a ailable
and he c edibili y e en o dia y-based es ima es emains ques ionable, we apply a
di e en app oach o s udy how social no ms a ec he epo ing in in e iews. Using
da a om he Ge man Socio-Economic Panel, we s udy how e ospec i e esponses
abou ime-use in ace- o- ace in e iews a e in luenced by he gende o he in e -
iewe . I social no ms play a ole in he epo ing o ime-use, i seems likely ha
he cha ac e is ics o he in e iewe , especially he in e iewe ’s gende , a ec in
which di ec ion social no ms in luence he ime-use es ima es and how s ongly. We
ocus on ou ine household cho es such as washing, cooking and cleaning, as hey a e
conside ed among he leas enjoyable and also he mos ime consuming non-ma ke
ac i i ies (Sulli an 2013; Col ane 2000), which poin s o he impo ance o social
no ms in explaining pe sis en gende di e ences in ime spen on hese asks. Fo
o he ac i i ies he connec ion o social no ms is less s aigh o wa d.
We ind ha emale esponden s epo signi ican ly mo e hou s o housewo k when
in e iewed by a woman ins ead o a man. A simila gende -o -in e iewe e ec o
male esponden s is no obse able. This in u n leads o an in e iewe gende gap in
he housewo k gende gap, because women and men espond di e en ly o he in e -
iewe ’s gende . We in e p e he indings as e idence o he ac ha social desi abili y
conce ns – and hence social no ms – play an impo an ole in he epo ing o house-
wo k hou s. By compa ing couples o singles, and also single- o-couple ansi ions o e
ime, we disen angle he mechanisms behind socially desi able epo ing, namely sub-
conscious sel -decep ion e sus imp ession managemen . S a i ying he sample and
in e ac ing he in e iewe ’s gende wi h se e al no m- ela ed cha ac e is ics, includ-
ing esponden s’ and in e iewe s’ coho s, suppo he hypo hesis ha he obse ed
e ec s a e ela ed o adi ional social no ms. We analyze di e ences by women’s el-
a i e income and employmen s a us and he eby discuss how he in e iewe ’s gende
is ela ed o he inding by Be and e al. (2015) ha couples “do gende ” when women
ou ea n hei pa ne s. Following he li e a u e on di e ences be ween Wes and Eas
Ge mans ega ding social no ms (Alesina and Fuchs-Schündeln 2007; Baue nschus e
123
2980 M.Ach ,M.Rebaudo
and Raine 2012; Beblo and Gö ges 2018), we ind ha gende -o -in e iewe e ec s
a e less p esen o Eas Ge mans. Fu he mo e, we s udy di e ences in he gende -
o -in e iewe e ec by sou ce coun y gende equali y o immig an s based on he
li e a u e on social no ms and cul u al he i age (Alesina e al. 2013; Blau e al. 2020;
Giuliano 2020). In line wi h ou hypo hesis, he e ec s a e smalle o immig an s
om mo e gende -equal coun ies.
We p o ide he i s causal e idence o how housewo k epo s di e by he in e -
iewe ’s gende , and how his gende -o -in e iewe e ec is ela ed o social no ms.
The e o e, ou pape con ibu es o bo h he s and o li e a u e on in e iewe e ec s
and ha on no ms a ec ing he di ision o housewo k hou s. We ind di e ences in
epo ing beha io be ween emale and male esponden s, indica ing ha gende gap
es ima es based on ace- o- ace in e iews should be in e p e ed wi h cau ion. We
a gue ha he ex en o gende ed epo ing is di ec ly ela ed o he in e naliza ion
o adi ional social no ms. The eby, we o e a new pe spec i e on p e ious indings
abou he ela ionship be ween social no ms and ime spen on housewo k, as hese
may be in luenced by epo ing e ec s. While ou app oach ocuses on epo ing
di e ences owa d emale e sus male in e iewe s, ou indings sugges ha social
no ms may in luence esponse beha io in e en mo e ways han could be analyzed in
his s udy.
The pape is s uc u ed as ollows. Sec ion 2summa izes he exis ing e idence
on gende -o -in e iewe e ec s and p o ides a heo e ical o e iew o he concep
o social desi abili y bias o explain he indings. Sec ion3desc ibes he da a and
Sec .4p esen s he empi ical s a egy. Sec ion 5s udies gende -o -in e iewe e ec s
on housewo k hou s and ela es he indings o se e al no m- ela ed cha ac e is ics,
including coho s, employmen s a us and income, di e ences be ween Wes and Eas
Ge many and he cul u al he i age o immig an s. Sec ion6p o ides obus ness checks
o add ess i any o he ac o s o he han he in e iewe ’s gende a e d i ing he esul s,
be o e Sec .7concludes.
2 Backg ound
Be o e summa izing he exis ing li e a u e on gende -o -in e iewe e ec s, we i s
p o ide a heo e ical o e iew o he concep o social desi abili y bias, which p o-
ides he ounda ion o he in e p e a ion o ou esul s. The concep o socially
desi able epo ing sugges s ha esponden s e lec on socie al iews and expec a-
ions ega ding ce ain opics and adap hei esponses acco dingly (see K umpal
2013, o an o e iew). Hence, ins ead o answe ing accu a ely and u h ully, some
esponden s dis o hei answe s o con o m o social no ms and he eby main ain a
socially a o able sel -p esen a ion. In ac , empi ical s udies ha e shown ha espon-
den s o e epo socially desi able beha io , such as cha i y dona ions, o ing, o sea
bel use, and unde epo socially undesi able beha io , such as smoking o illici d ug
use (Tou angeau e al. 2000; Tou angeau and Yan 2007; K umpal 2013).
The unde lying p ocesses and exac mechanisms o socially desi able epo ing a e
s ill insu icien ly unde s ood (Tou angeau and Yan 2007; Hol g a es 2004). Schola s
o en dis inguish be ween wo ypes o no m-con o ming epo ing: O he -decep ion
123
A gende gap in gende gaps… 2981
o imp ession managemen , on he one hand, ela es o he indi iduals’ need o social
app o al by o he s, e.g., he in e iewe (Paulhus 1991). I assumes ha some espon-
den s pu pose ully ailo hei answe s o a oid social disapp o al o o c ea e a posi i e
image o hemsel es. Sel -decep ion, on he o he hand, e e s o indi iduals’ s a egies
o educe cogni i e dissonance caused by a di e gence be ween social no ms and eal-
i y (K umpal 2013). As opposed o imp ession managemen , sel -decep ion is mos ly
conside ed o be subconscious and au oma ic (Tou angeau and Yan 2007). While
imp ession managemen beha io is di ec ed a o he subjec s, e.g., in e iewe s, he
main add essees o sel -decep i e epo ing beha io a e he esponden s hemsel es.
Impo an ly, independen o he channel, socially desi able epo ing will only occu
when a de ia ion be ween he hones esponse and wha is pe cei ed o be desi able
exis s.
I has been shown ha social desi abili y bias is especially likely in pe sonal
in e iews and ha in e iewe cha ac e is ics, such as he in e iewe ’s gende , can
a ec he likelihood o engaging in no m-con o ming epo ing beha io (Tou angeau
e al. 2000; Tou angeau and Yan 2007). In e ms o imp ession managemen , espon-
den s may pe cei e emale and male in e iewe s o ha e di e en opinions ega ding
gende - ela ed ques ions and adjus hei esponses o gain social app o al o he in e -
iewe . Howe e , e en i esponden s we e no ying o imp ess hei coun e pa s, he
in e iewe ’s gende can s ill a ec he esponse in e ms o sel -decep i e beha io .
Speci ically, esponden s may be mo e o less awa e o hei own de ia ion om social
no ms depending on hei coun e pa ’s gende .
Following his concep o socially desi able epo ing, se e al s udies analyze he
ole o gende -o -in e iewe e ec s (see Wes and Blom 2017, o an o e iew).
Mos o he p e ious esea ch has ocused on no ma i e iews, o example abou
ma iage (Liu and S ainback 2013) o women’s issues and gende equali y (Huddy
e al. 1997). A common inding in he li e a u e is ha esponden s ei he p o ide mo e
eminis o libe al esponses owa d emale in e iewe s o , he o he way a ound, mo e
adi ional ones owa d male in e iewe s (Kane and Macaulay 1993; Luep ow e al.
1990; Flo es-Macias and Lawson 2008; Lipps and Lu z 2017).
I is impo an o no e ha p e ious s udies on gende -o -in e iewe e ec s almos
exclusi ely ely on c oss-sec ional da a and hence canno es ablish causal e ec s.
Fu he mo e, sample sizes a e usually small and only a ew in e iewe s conduc he
su eys (Huddy e al. 1997). A majo disad an age o c oss-sec ional s udies in his
con ex is ha in e iewe e ec s a e po en ially con ounded wi h a ea e ec s, as in e -
iewe s a e ypically alloca ed o speci ic geog aphic a eas (Schnell and K eu e 2005).
To ou knowledge, only wo s udies e alua ing gende -o -in e iewe e ec s ela ed
o gende issues add ess causali y conce ns by analyzing panel da a, Zoch (2021) and
Lipps and Lu z (2017). Zoch (2021) uses da a om he Na ional Educa ional Panel
S udy in Ge many and analyzes how he in e iewe ’s gende a ec s sel - epo ed
gende ideologies. In line wi h p e ious esea ch on gende issues, she inds ha
esponden s epo less adi ional a i udes owa d emale in e iewe s. Acco ding o
his s udy, he e a e, howe e , no signi ican gende -o -in e iewe e ec s o an a i-
udinal i em asking whe he i ’s he man’s job o ea n money and he woman’s job o
ake ca e o he household and amily. Housewo k hou s o o he ime-use ca ego ies
a e no analyzed. The s udy by Lipps and Lu z (2017) is based on Swiss panel da a
123
2982 M.Ach ,M.Rebaudo
om elephone in e iews and e alua es gende -o -in e iewe e ec s on a a ie y o
di e en opics. Some o he analyzed i ems a e ela ed o ac ual household asks
(e.g., “in ou household, i is mos ly me who does he cleaning”) and he au ho s also
analyze hou s o housewo k bu do no ind signi ican gende -o -in e iewe e ec s
o any o hese i ems. Because i has a di e en ocus, he s udy does no analyze how
esul s may di e by subg oups o o he no m- ela ed cha ac e is ics. I is he e o e
possible ha po en ial gende -o -in e iewe e ec s a e hidden, o example, because
in ou s udy we only ind e ec s o couples and no o singles. Fu he mo e, he
di ec in e ac ion wi h an in e iewe (as opposed o a elephone in e iew) is likely
heigh ening he salience o no ms which we a gue o d i e he e ec s.
Ou esea ch ques ion ocuses on a ac ual su ey ques ion – ime spen on house-
wo k – and hence di e s om he majo i y o o he s udies abou gende -o -in e iewe
e ec s ha analyze a i udinal ques ions. Apa om Lipps and Lu z (2017) who do
no ind e ec s, he e a e no o he s udies e alua ing gende -o -in e iewe e ec s on
housewo k hou s. Impo an ly, housewo k hou s a e po en ially e y di e en om
a i udinal ques ions abou gende equali y. In pa icula , he ela ionship be ween
hou s o housewo k and social no ms is mo e sub le. While esponden s may engage
in imp ession managemen when answe ing a i udinal ques ions on gende equali y
o ea o o ending he in e iewe , his seems unlikely ega ding he esponse abou
housewo k hou s. Ins ead, we a gue ha he u ge o ul ill a pa icula gende ole is
subconsciously a ec ed by he in e iewe ’s p esence. Fo example, esponden s may
become mo e o less awa e o hei own de ia ion om social no ms depending on
he in e iewe ’s gende . In line wi h sel -decep i e epo ing beha io his pe cep ion
may subsequen ly in luence he hough p ocess o inding and o mula ing an answe ,
gi en ha mos people a e unce ain abou hei exac housewo k hou s. Ou empi i-
cal analysis o di e en subg oups helps o disen angle he possible mechanisms o
socially desi able epo ing and suppo s his hypo hesis o sel -decep i e beha io .
3 Da a
We use da a om he Ge man Socio-Economic Panel (SOEP), a longi udinal su -
ey o abou 15,000 p i a e households in Ge many wi h annual in e iews since
1984 (Goebel e al. 2019).1The majo i y o in e iews in he SOEP a e ace- o- ace
in e iews, which allows us o s udy he e ec o he in e iewe ’s gende on indi-
idual esponses. No ably, SOEP in e iews a e highly s anda dized. In e iewe s
a e employed by he ield ins i u e Kan a , which p io i izes in e iewe moni o ing
h ough ISO-ce i ied p ocesses ha a e egula ly audi ed (Bohlende e al. 2020).
Fo ins ance, a leas 10% o annual in e iews a e checked, and each in e iewe is
moni o ed a leas once a yea . In e iewe s a e ained o ensu e high da a quali y.
They pa icipa e in annual aining sessions conduc ed by he SOEP eam a Kan a ,
along wi h ep esen a i es om DIW Be lin, whe e SOEP is based. These sessions
p epa e con ac in e iewe s, who hen ain in e iewe s in he egions o which hey
a e esponsible. Con en s o hese ainings a e scope, iming, and p ocedu es o he
1Socio-Economic Panel (SOEP), da a o yea s 1984-2018, e sion 35, 2020.
123
A gende gap in gende gaps… 2983
in e iews, as well as any special ea u es speci ic o each su ey yea (Goebel e al.
2019; Bohlende e al. 2020).
Ou main ou come o in e es is he ime spen on housewo k, which is measu ed
using he ollowing ques ion (see he en i e ques ion in Fig.1in he Appendix): “Wha
is a ypical weekday like o you? How many hou s pe no mal wo kday do you spend
on housewo k (washing, cooking, cleaning)?”. This de ini ion o housewo k aligns
wi h he ypical usage in he li e a u e (e.g., Lippmann e al. 2020), speci ically exclud-
ing o he non-ma ke ac i i ies like childca e o epai wo k and ga dening, which a e
lis ed as sepa a e ac i i ies in he same in e iew ques ion. Rou ine housewo k is con-
side ed by bo h women and men as one o he leas enjoyable ac i i ies (Ge shuny
2013; Sulli an 2013), making i an in e es ing subjec o s udy in he con ex o wi hin-
couple di e ences. Tasks like epai wo k and ga dening, unlike ou ine housewo k,
ha e been ound o be mo e ime lexible and mo e enjoyable (Col ane 2000). As
a gued by Sulli an (2010), he no m conce ning childca e has shi ed since he 1970s,
wi h ime spen wi h child en inc easingly being iewed as leisu e, pa icula ly among
mo e educa ed pa en s. Gi en ha ou ine housewo k is mos di ec ly ela ed o social
no ms, we es ic ou ocus on housewo k hou s o he main speci ica ion. How-
e e , we also p o ide obus ness checks o he o he ac i i ies lis ed in he ime-use
ques ion.
Ou analysis co e s all yea s om 1985 o 2018, as in o ma ion abou he in e iew-
e s is a ailable om 1985 onwa ds. We ollow Lippmann e al. (2020) and Be and
e al. (2015), and ocus on wo king-age indi iduals be ween 18 and 65 yea s o age,
since we can analyze he ole o employmen s a us and ( ela i e) income o his
g oup. The main sample consis s o cohabi ing couples (ma ied o no ma ied),
al hough addi ional esul s o singles will be discussed in sec ion 5. Impo an ly, in
he SOEP da a, all adul s in he household answe he pe sonal ques ionnai e, including
he ime-use epo s, hemsel es. The eby, wi hin-couple gaps in housewo k epo s
can be analyzed. We exclude he ew obse a ions o pa ne s who we e in e iewed
by di e en in e iewe s.2We u he mo e exclude mo e-gene a ion-households and
o he non- ypical household a angemen s. Because he in e iewe ’s gende may di -
e en ly a ec homosexual indi iduals, we also exclude hem om he analysis.3Ou
inal sample consis s o 18,965 couples ha a e obse ed o an a e age o a ound six
yea s. Table 1p esen s desc ip i e s a is ics o he main sample. While women spend
on a e age abou h ee hou s pe weekday on housewo k du ies, he a e age o men
is 46min pe day.4O all indi iduals conside ed (employed and unemployed), he
a e age wo king ime o women is nea ly hal ha o men, and incomes a e consis-
en ly lowe . Men in ou sample a e a bi olde and on a e age ha e a sligh ly highe
2In he inal sample his conce ns only 68 obse a ions. Couples a e no excluded i pa ne s a e in e iewed
on di e en da es. In ou sample 88% o couples a e in e iewed on he same day and a ound 95% wi hin
se en days a e he i s pa ne ’s in e iew. Resul s a e almos iden ical when excluding couples who we e
in e iewed on di e en da es, see Table 11 in he Appendix.
3In ou sample, he e a e only 112 di e en same-sex couples.
4In ou sample women’s maximum housewo k ime is 24h and men’s is 16h pe day. When we assume
8h o sleep, housewo k es ima es o mo e han 16h a e likely o e s a ed. Howe e , only 32 obse a ions
epo emale housewo k hou s o be la ge han 16. Ou esul s a e obus when obse a ions epo ing
high le els o housewo k, such as mo e han 16 h o e en mo e han 6h, a e excluded, see Table 12 in he
Appendix.
123
2984 M.Ach ,M.Rebaudo
Table 1 Desc ip i e s a is ics
Mean S anda d de ia ion Min Max
Woman’s housewo k ime 3.00 1.95 0 24
Man’s housewo k ime 0.76 0.93 0 16
Housewo k woman - housewo k man 2.24 2.23 −15 24
Paid wo k ime woman 19.47 18.08 0 80
Paid wo k ime man 36.69 18.69 0 80
Income woman 775.25 908.06 0 36184
Income man 2054.58 1725.40 0 98038
Woman’s age 42.58 10.72 18 65
Man’s age 45.33 10.77 18 65
Educa ion le el woman 4.92 2.33 0 9
Educa ion le el man 5.01 2.43 0 9
Ma ied 0.88 0.33 0 1
No. o child en age <6 0.30 0.59 0 6
No. o child en age 6-12 0.43 0.73 0 5
No. o child en age 13-16 0.23 0.50 0 4
Income HH 3447.94 2062.53 43 210084
Eas Ge many 0.21 0.41 0 1
Obse a ions 110,544
The able epo s summa y s a is ics o esponden s’ cha ac e is ics based on he main sample (cohabi ing
couples o age 18 o 65). Housewo k ime is hou s pe day. Paid wo k ime is hou s wo ked pe week
including o e ime hou s. Income is ne mon hly income and household income is ne mon hly income
o all household membe s including non-labo income. Income is co ec ed o in la ion wi h base yea
2016. Educa ion le els de i e om he CASMIN classi ica ion. Ma ied is a dummy a iable equal o 1
o ma ied couples. No. o child en is he numbe o child en li ing in he household o each o he age
g oups. Eas Ge many is a dummy a iable equal o 1 when he couple li es in Eas Ge many a he ime o
he su ey
educa ion le el. 88% o he couples obse ed a e ma ied and a ound 21% li e in Eas
Ge many a he ime o he su ey.
In o al, ou sample consis s o 1,841 di e en in e iewe s. A ound 57% o all
in e iews a e pe o med by male in e iewe s. Female and male in e iewe s di e
in some cha ac e is ics ha may a ec in e iew esponses, as shown in Table 2.Male
in e iewe s on a e age ha e sligh ly less expe ience wi h SOEP, a e olde , mo e o en
ma ied, mo e likely o ha e a highe educa ion le el, mo e o en bo n in Ge many, and
mo e o en o Ge man mo he ongue. Un o una ely, mos ques ions abou addi ional
in e iewe cha ac e is ics, excep he in e iewe ’s age and expe ience, a e only asked
in speci ic yea s, so ha including hem as con ols in he eg ession analyses esul s
in a mo e selec i e sample. The e o e, we only include he in e iewe ’s gende and
age o he main analysis and p o ide speci ica ions wi h he addi ional cha ac e is-
ics as obus ness checks in sec ion 6.5Apa om he obse ed di e ences be ween
5In e iewe ’s expe ience does no a ec housewo k epo s and he gende -o -in e iewe e ec is una -
ec ed when addi ionally con olling o expe ience, see Table 13 in he Appendix. We do no include i as
123
A gende gap in gende gaps… 2991
he es ima ed coe icien s, being in a couple ela ionship as opposed o being single,
o women esul s in mo e housewo k hou s, while o men i esul s in less. Mo e
in e es ingly o ou pu pose, he in e ac ion e ec be ween he Couple dummy and
he in e iewe ’s gende o women is s a is ically signi ican and posi i e. Hence, he
es ablished esul ha he gende -o -in e iewe e ec is p esen o women in couple
ela ionships bu no o singles can also be ound when s udying ansi ions wi hin
indi iduals o e ime. As expec ed om he p e ious esul s he e is no s a is ically
signi ican in e ac ion e ec o men.
The compa ison o couples and singles helps o u he disen angle he p ecise
mechanisms behind socially desi able epo ing, namely sel -decep ion e sus imp es-
sion managemen . Abo e, we ha e a gued ha i seems unlikely o esponden s o
ea o ending he in e iewe , as housewo k hou s a e di e en om a i udinal i ems
on gende equali y. The e a e wo addi ional explana ions consis en wi h imp ession
managemen as o why esponden s may adjus hei housewo k epo s based on he
gende o he in e iewe . Fi s ly, indings on expe imen e -gende -e ec s om o he
sciences sugges ha he e migh be a gene al desi e o imp ess one’s coun e pa ,
especially in opposi e-sex ela ionships ( o he e osexual indi iduals) (Chapman e al.
2018). Indi iduals wan o be pe cei ed posi i ely by he opposi e sex, which can lead
o sel -a i ming eedback o e en he possibili y o a oman ic ela ionship. In ela-
ion o ou esul s, his mechanism could indeed play a ole i women we e epo ing
mo e housewo k owa d male in e iewe s, bu i seems implausible ha epo ing
less owa d male in e iewe s would imp o e women’s pe cei ed a ac i eness as
po en ial pa ne s. Fu he mo e, i such a channel we e in place, i should also be
appa en o single women, since a ac i eness on he ma iage ma ke is likely e en
mo e ele an o hem. Hence, bo h he di ec ion o he e ec and he non-exis ence
o e ec s o single women con adic such a channel.
Secondly, and also in line wi h imp ession managemen heo y, esponden s may
ha e a gene al desi e o no being conside ed as unclean by o he indi iduals when
hei own housewo k hou s a e pe cei ed oo low. I women expec highe cleanliness
s anda ds om o he women as opposed o men, his channel could be in line wi h
highe housewo k epo s owa d emale in e iewe s. Howe e , i hose conce ns we e
indeed d i ing he esul s, again one would expec simila e ec s o single women.
As we do no ind hose, gene al expec a ions ega ding cleanliness s anda ds also
do no seem o explain ou esul s. As a consequence, subconscious sel -decep ion
ins ead o imp ession managemen seems o be he mo e ele an channel o explain
he obse ed e ec .
By compa ing couples e sus singles, we compa e gende -o -in e iewe e ec s
o subg oups which also di e wi h espec o hei o al ime spen on housewo k.
The e o e, we need o add ess he issue o absolu e e sus ela i e adjus men s in he
housewo k epo . Fo ins ance, i he ex en o gende ed epo ing depended on he
baseline le el o housewo k, he absolu e gende -o -in e iewe e ec migh only be
small and insigni ican o singles because hey do less housewo k han women in
couple ela ionships. The e o e, in Table 19 in he Appendix we analyze he gende -
o -in e iewe e ec in ela i e e ms by using he loga i hm o housewo k hou s as he
dependen a iable. The esul s a e quali a i ely and quan i a i ely in line wi h he main
123
2992 M.Ach ,M.Rebaudo
Table 4 Housewo k ime and in e iewe gende - singles
Singles Single-couple
Women Men Women Men
(1) (2) (3) (4)
Female in e iewe −0.046 0.018 −0.005 −0.005
(0.03) (0.03) (0.03) (0.02)
Couple 0.190*** −0.396***
(0.03) (0.02)
Female in e iewe * couple 0.073** 0.023
(0.03) (0.03)
Adjus ed R20.101 0.054 0.171 0.088
Indi iduals 7564 4783 23668 21486
Obse a ions 33912 20334 144456 130878
Dependen a iable is emale o male daily housewo k hou s. The able epo s he esul s o indi idual
ixed e ec s es ima ions. Responden s and In e iewe con ols as well as yea and s a e ixed e ec s a e
included. S anda d e o s clus e ed a he indi idual le el a e in pa en heses. Signi icance is deno ed by
*p<.10, ** p<.05, *** p<.01
speci ica ion. Impo an ly, e en in ela i e e ms he e a e no gende -o -in e iewe
e ec s o men o singles.
To summa ize, women in couple ela ionships epo mo e housewo k hou s
owa d emale in e iewe s which we in e p e as e idence o social no ms a ec ing
housewo k epo s. Following he heo y on social desi abili y, we expec gende -o -
in e iewe e ec s o be la ge when he de ia ion be ween wha is pe cei ed o be
desi able and eali y is la ge . Mo eo e , gende -o -in e iewe e ec s may depend on
he in e naliza ion o adi ional social no ms. These hypo heses will be examined in
he subsequen sec ions.
5.3 Gende ed epo ing and no m- ela ed cha ac e is ics
I he epo ing o housewo k hou s is d i en by he desi e o ul ill a ce ain ole, he
size o he e ec may di e by di e en no m- ela ed cha ac e is ics. S udies show
ha social no ms can e ol e o e ime (Fe nández 2013) so ha la e -bo n gene a ions
possibly hold less adi ional iews abou he ole o women and men. Wi h ega d
o ou esea ch ques ion, i is he e o e in e es ing o analyze whe he he gende -o -
in e iewe e ec di e s by esponden and in e iewe coho s.
Table 5p esen s he esul s. We spli esponden s up in o wo g oups by he h esh-
old o 1960 as yea o bi h, which is oughly he sample mean. As columns 1 and 2
show, he gende -o -in e iewe coe icien indeed appea s o be smalle o he la e
bo n coho o women, despi e s ill being ma ginally signi ican . In column 3 we ana-
lyze how he e ec di e s by coho s o in e iewe s. Wi h esponden ixed-e ec s
models we can obse e a ia ion in he in e iewe ’s coho o e ime (bu no in he
esponden s’ coho s), so we spli in e iewe s up in o h ee ins ead o wo g oups. In
123
A gende gap in gende gaps… 2993
Table 5 Women’s housewo k ime - esponden and in e iewe coho s
Young coho Old coho All Young coho Old coho
(1) (2) (3) (4) (5)
Female in e iewe 0.049* 0.117*** 0.103*** 0.078* 0.121**
(0.03) (0.04) (0.03) (0.04) (0.05)
[Re e ence: In e iewe bo n <1945]
In e iewe bo n 1945-1959 −0.001 0.044 −0.063
(0.05) (0.05) (0.08)
In e iewe bo n >=1960 −0.017 −0.022 0.040
(0.08) (0.09) (0.14)
Female in e iewe −0.015 −0.026 0.029
* In e iewe bo n 1945-1959 (0.05) (0.07) (0.08)
Female in e iewe −0.168** −0.150* −0.163
* In e iewe bo n >=1960 (0.08) (0.08) (0.16)
Adjus ed R20.156 0.184 0.167 0.156 0.184
Indi iduals 12027 6938 18965 12027 6938
Obse a ions 63757 46787 110544 63757 46787
Dependen a iable is emale daily housewo k hou s. The sample is es ic ed o cohabi ing couples (age 18 o 65). Columns a e based on esponden coho s: Young coho is
de ined as a leas one pe son in he couple being bo n 1960 o la e . The able epo s he esul s o indi idual ixed e ec s es ima ions. Responden s and In e iewe con ols
as well as yea and s a e ixed e ec s a e included as in he main speci ica ion. S anda d e o s clus e ed a he indi idual le el a e in pa en heses. Signi icance is deno ed by
*p<.10, ** p<.05, *** p<.01
123
2994 M.Ach ,M.Rebaudo
addi ion o he 1960- h eshold ha is compa able o esponden s’ coho s, we sepa-
a ely accoun o in e iewe s bo n be o e 1945 and in e iewe s bo n be ween 1945
and 1959. The gende -o -in e iewe coe icien pe sis s owa d in e iewe s o he
wo olde coho s, which ep esen mo e han 90% o in e iewe s in ou sample. How-
e e , he e is a e e se e ec o he younges coho o in e iewe s, esul ing in an
insigni ican gende -o -in e iewe coe icien o esponden s being in e iewed by
hese in e iewe s. When we compa e he e ec s o he in e iewe coho s by he wo
esponden coho s (columns 4 and 5), i appea s ha especially esponden s om he
younge coho engage less in gende ed epo ing when in e iewed by he younges
gene a ion o in e iewe s, howe e , he e may also be mo e ansi ions owa d young
in e iewe s in his g oup. The esul ing gende -o -in e iewe coe icien o he
young esponden coho owa d young in e iewe s is no s a is ically di e en om
ze o. Fo male esponden s, we s ill do no ind any gende -o -in e iewe e ec s
when sepa a ely s udying esul s by coho s (see Table 20 in he Appendix). Sum-
ma izing, he e is some e idence ha gende -o -in e iewe epo ing is less p esen
when esponden s and mo e impo an ly in e iewe s a e om younge coho s and
hence may hold less adi ional iews. This suppo s ou hypo hesis ha gende ed
epo ing is indeed ela ed o social no ms. We do no ind ha he in e iewe ’s age
(independen o he bi h coho ) is ela ed o he gende -o -in e iewe e ec (see
Table 21 in he Appendix).
Fu he mo e, we analyze how esul s di e by educa ion le els and by he p esence
o young child en, as bo h may be ela ed o he ole esponden s wish o ul ill. Fo
example, mo he s o small child en may ha e a s onge u ge o be pe cei ed as “good”
mo he s by epo ing mo e housewo k hou s. Highly educa ed women, in con as , may
eel less p essu e o con o m o adi ional oles. In e es ingly, we ind ha he gende -
o -in e iewe e ec does no di e be ween esponden s o low o high educa ion
and also no by he p esence o child en in he household (see Tables 22 and 23 in
he Appendix). This u he co obo a es ha gende -o -in e iewe e ec s no only
depend on esponden s’ pe cei ed social no ms, bu also on he de ia ion be ween
he hones esponse and wha is pe cei ed o be desi able. By child en and educa ion
s a us, such a de ia ion does no seem o di e much, despi e po en ial di e ences
ega ding pe cei ed social no ms.
5.4 Gende ed epo ing and specializa ion in paid wo k
I he epo ing o housewo k hou s is d i en by he desi e o ul ill a ce ain ole, he
size o he e ec may di e by women’s employmen s a us and income. We suppose
ha a woman’s employmen s a us (in addi ion o he wo king hou s) may a ec he
belie abou which social ole o ul ill: Fo example, ull- ime employed women –
compa ed o o he women – migh eel less p essu e o con o m o he housewi e
ole and hence eel less u ge o (subconsciously) adjus housewo k epo s. Then, he
gende -o -in e iewe e ec ha po en ially cap u es he need o ul ill socie al expec-
a ions, would be smalle o ull- ime employed women. Acco dingly, a woman’s
con ibu ion o he couple’s income migh also a ec he belie abou socie al expec-
a ions ega ding housewo k. In line wi h he p e ious hypo hesis, i could be ha
123
A gende gap in gende gaps… 2995
women wi h a highe con ibu ion o household income eel less need o adjus house-
wo k epo s o ul ill he housewi e ole because hey a e iden i ying hemsel es
mo e as wo king women. In con as , he “doing gende ” li e a u e would sugges ha
a woman ea ning mo e han he pa ne eels he need o con o m o he adi ional
ole by ei he doing mo e housewo k o , po en ially, epo ing mo e housewo k hou s
(Lippmann e al. 2020; Be and e al. 2015). Acco ding o his idea, women who ou -
ea n hei male pa ne s migh show a s onge gende -o -in e iewe e ec because
o he desi e o con o m o hei social ole.
To es hese hypo heses, we add women’s employmen s a us and ela i e income
as con ols o he eg ession analyses. Resul s a e p esen ed in Table 6. Fo hese
eg essions, we es ic he sample o dual-ea ne couples as he in e p e a ion o
di e ences in ela i e income and employmen s a us is only meaning ul o his g oup.
In con as o ela i e income, we de ine he ull- ime dummy in absolu e e ms o
emales (no in ela i e e ms wi hin he couple) because in ou sample o dual-ea ne
couples, 95% o men wo k ull- ime bu only 43% o women do. Whe he he woman
is wo king ull- ime and he ela i e income a e bo h included in he eg essions as
in e ac ion e ms wi h he in e iewe ’s gende and also used o subg oup analyses.
Consis en wi h ou i s hypo hesis, unning sepa a e eg essions o ull- ime and
no - ull- ime employed (pa - ime/i egula ly employed) women (columns 1 and 2)
shows ha he gende -o -in e iewe e ec o he la e is almos wice as la ge as
o he o me . Hence, i seems o be he case ha ull- ime employed women eel
less need o adjus housewo k epo s. Including women’s employmen s a us as an
in e ac ion e ec wi h he in e iewe ’s gende (column 3) sugges s ha ull- ime
employmen gene ally educes housewo k hou s, bu he ela ionship be ween ull-
ime employmen and he in e iewe ’s gende is no s a is ically signi ican .
To s udy he ole o ela i e income we de ine wo a iables, ollowing Be and e al.
(2015). Fi s ly, he a iable Rela i eIncomeWoman measu es he woman’s sha e in
he couple’s income, de ined as IncomeWoman/(IncomeWoman+IncomeMan).
Secondly, we de ine a dummy a iable WomanEa nsMo e coded one when he
woman ea ns s ic ly mo e han he pa ne , and ze o o he wise. When sepa a ely
s udying couples whe e he man ea ns mo e (o incomes a e equal) and couples whe e
he woman ea ns mo e (columns 4 and 5), we see ha he gende -o -in e iewe e ec
is smalle and no s a is ically signi ican in he la e g oup. This suppo s he i s
hypo hesis ha women wi h a highe con ibu ion o he couple’s income eel less need
o adjus hei housewo k epo s o ul ill a ce ain ole. Fu he mo e, we un wo addi-
ional eg essions on he whole sample: In column 6, we con ol o women’s ela i e
income and addi ionally include an in e ac ion e ec be ween he in e iewe ’s gen-
de and a dummy a iable indica ing whe he he woman ea ns mo e han he man.10
The ela ionships shown by Be and e al. (2015) and Lippmann e al. (2020)a e
also p esen when addi ionally con olling o he in e iewe ’s gende : La ge ela-
i e income dec eases emale housewo k hou s, bu he e is a s a is ically signi ican
10 This speci ica ion is simila o he ones by Be and e al. (2015) and Lippmann e al. (2020)as hey
con ol o he woman’s income sha e and addi ionally o a dummy a iable indica ing whe he he woman
ea ns mo e han he man. They do so in o de o cap u e only he e ec associa ed wi h he iola ion o
he social no m ha “a man should ea n mo e han his wi e” i espec i e o specializa ion e ec s ela ed o
ela i e income.
123
2996 M.Ach ,M.Rebaudo
Table 6 Women’s housewo k ime and in e ac ions - dual-ea ne couples
Employmen S a us Rela i e Income
Full- ime No ull- ime All Male highe o equal Female highe All
(1) (2) (3) (4) (5) (6) (7)
Female in e iewe 0.058** 0.107*** 0.096*** 0.088*** 0.042 0.079*** 0.130***
(0.03) (0.03) (0.03) (0.03) (0.04) (0.02) (0.04)
Female in e iewe −0.047
* Woman ull- ime (0.03)
Woman ull- ime −0.121***
(0.03)
Female in e iewe −0.033
* Woman ea ns mo e (0.03)
Female in e iewe −0.167*
* Rela i e income woman (0.10)
Woman Ea ns Mo e 0.127*** 0.113***
(0.02) (0.02)
Rela i e income woman −1.148*** −1.077***
(0.10) (0.11)
Adjus ed R20.045 0.088 0.099 0.101 0.048 0.100 0.100
Indi iduals 7032 8890 13145 11861 3349 13145 13145
Obse a ions 26695 34962 61657 51779 9878 61657 61657
Dependen a iable is emale daily housewo k hou s. The sample is es ic ed o dual-ea ne couples (age 18 o 65). The able epo s he esul s o indi idual ixed e ec s
es ima ions. Responden s and In e iewe con ols as well as yea and s a e ixed e ec s a e included. S anda d e o s clus e ed a he indi idual le el a e in pa en heses.
Signi icance is deno ed by * p<.10, ** p<.05, *** p<.01
123
A gende gap in gende gaps… 2997
and posi i e coe icien o he WomanEa nsMo edummy, which ep esen s he so-
called doing gende e ec . The in e ac ion e ec o he in e iewe ’s gende wi h he
WomanEa nsMo e dummy in his eg ession is no s a is ically signi ican . When
he in e iewe ’s gende is in e ac ed wi h ela i e income (ins ead o he dummy
a iable) (column 7), we obse e a nega i e and ma ginally signi ican in e ac ion
e ec . Thus, he highe he woman’s con ibu ion o he couple’s income, he less hey
engage in gende ed epo ing. In bo h speci ica ions, he o e all gende -o -in e iewe
coe icien emains posi i e and s a is ically signi ican .
These esul s combined wi h he p e ious ones sugges ha women who a e ela-
i ely mo e a ached o paid wo k, eel less need o epo mo e housewo k hou s owa d
emale in e iewe s. This is in line wi h he expec a ion ha gende -o -in e iewe
e ec s will be la ge when he de ia ion be ween wha is expec ed o be desi able and
eali y is la ge . While we oucs on dual-ea ne couples o hese analyses, Table 24 in
he Appendix also e alua es whe he being employed ( s. no being employed) a ec s
he esul s. As shown, he gende -o -in e iewe e ec does no di e by employmen
s a us a he ex ensi e ma gin. As unemployed women ypically do mo e housewo k
han employed women, i is likely ha hey do no pe cei e a s onge de ia ion
be ween hei ac ual hou s and hei expec ed hou s, compa ed o wo king women.
We do no ind e idence o he second hypo hesis ha women ea ning mo e han
hei pa ne s eac mo e s ongly o he in e iewe ’s gende . No e, howe e , ha hese
esul s nei he con i m no con adic he gene al idea o “doing gende ” epo ing,
whe e women iola ing adi ional social no ms in he ma ke sphe e (by ou ea ning
hei pa ne s) epo mo e housewo k hou s o es o e hei gende iden i y. I could,
o example, s ill be he case ha hese women epo mo e housewo k hou s, bu do
so simila ly when con on ed wi h emale and male in e iewe s.
Besides employmen s a us and ela i e income, egional di e ences p esen
ano he ca ego y o no m-p oxies o es hei ela ion o gende -o -in e iewe e ec s:
Following Alesina and Fuchs-Schündeln (2007), se e al s udies exploi he na u al
expe imen cons i u ed by he 41-yea di ision o Ge many o s udy di e ences in
social no ms be ween Wes e n and Eas e n egions. Among s udies e alua ing gen-
de a i udes and no ms, esul s show ha Eas Ge mans end o hold mo e egali a ian
gende - ole a i udes han Wes Ge mans (Baue nschus e and Raine 2012; Beblo and
Gö ges 2018). Lippmann e al. (2020) analyze how he esul s p esen ed by Be and
e al. (2015) di e be ween Wes and Eas Ge many and ind subs an ial di e ences.
Speci ically, hey show ha “doing gende ” beha io , i. e. he u ge o compensa e
o iola ions o he “male b eadwinne ” no m (by manipula ing labo ma ke ou -
comes o by adjus ing housewo k hou s), is p esen o Wes Ge mans bu no o
Eas Ge mans. They explain he di e ences ia he long-las ing e ec s o he mo e
gende -equal Eas Ge man ins i u ions. Following he inding ha Eas Ge mans hold
less adi ional iews ega ding he gende di ision o paid and unpaid wo k, i is
in e es ing o e alua e whe he and how he gende -o -in e iewe e ec migh di e
be ween Wes and Eas Ge mans.11 The esul s a e p esen ed in Table 7. Sepa a ely
11 Eas Ge many he e is de ined as he cu en egion o esidence a he ime o he su ey and includes Eas
Be lin bu no Wes Be lin. Due o he geog aphical de ini ion he e is he possibili y o wi hin- esponden
a ia ion on he Eas dummy. Al e na i ely, Eas can be de ined based on he biog aphical in o ma ion
abou he esidence egion be o e 1989. Resul s a e simila o his de ini ion and a ailable upon eques .
123
2998 M.Ach ,M.Rebaudo
Table 7 Women’s housewo k ime—Wes e sus Eas Ge many
Wes Eas All
(1) (2) (3) (4) (5)
Female in e iewe 0.077*** 0.058 0.078*** 0.105*** 0.140***
(0.03) (0.04) (0.03) (0.03) (0.05)
Eas 0.374 0.252 0.094
(0.26) (0.27) (0.29)
Female in e iewe * Eas −0.011 −0.064 −0.077
(0.04) (0.06) (0.10)
Woman ull- ime −0.166***
(0.03)
Woman ull- ime * Eas 0.142***
(0.05)
Female in e iewe * woman ull- ime −0.070*
(0.04)
Female in e iewe * woman ull- ime * Eas 0.111*
(0.06)
Rela i e income woman −1.235***
(0.13)
123
A gende gap in gende gaps… 2999
Table 7 con inued
Wes Eas All
(1) (2) (3) (4) (5)
Rela i e income woman * Eas 0.597***
(0.17)
Female in e iewe * ela i e income woman −0.204*
(0.12)
Female in e iewe * Eas * ela i e income woman 0.211
(0.21)
Adjus ed R20.107 0.053 0.098 0.099 0.101
Indi iduals 10462 2748 13145 13145 13145
Obse a ions 47792 13865 61657 61657 61657
Dependen a iable is emale daily housewo k hou s. The sample is es ic ed o dual-ea ne couples (age 18 o 65). The able epo s he esul s o indi idual ixed e ec s
es ima ions. Responden s and In e iewe con ols as well as yea and s a e ixed e ec s a e included. S anda d e o s clus e ed a he indi idual le el a e in pa en heses.
Signi icance is deno ed by * p<.10, ** p<.05, *** p<.01
123
3000 M.Ach ,M.Rebaudo
s udying Eas Ge man and Wes Ge man women as in columns 1 and 2, sugges s ha
he gende -o -in e iewe e ec is smalle and no s a is ically signi ican o Eas
Ge man women. The in e ac ion e ec be ween he in e iewe ’s gende and Eas o
he whole sample is no s a is ically signi ican .
To gain u he insigh in o di e ences be ween he g oups, we e alua e how he ela-
ionship be ween ull- ime employmen o ela i e income and he in e iewe ’s gende
di e s be ween Eas and Wes Ge mans. Speci ically, we include a h ee-way in e ac-
ion o he in e iewe ’s gende , he Eas dummy and women’s ull- ime employmen
in column 4 and women’s ela i e income in column 5. In bo h speci ica ions, he
gene al gende -o -in e iewe e ec emains s a is ically signi ican and he e is no
s a is ically signi ican in e ac ion e ec be ween Eas and FemaleIn e iewe .As
shown be o e in Table 6, ull- ime employed women gene ally do less housewo k, o
Eas Ge man women, howe e , he e is a e e se posi i e in e ac ion e ec ha is s a-
is ically signi ican . Mo e in e es ing o ou pu pose a e he in e ac ion e ec s wi h
he in e iewe ’s gende : While he in e ac ion be ween he in e iewe ’s gende and
women’s ull- ime employmen is s ill nega i e and now e en ma ginally signi ican ,
he h ee-way in e ac ion wi h women’s ull- ime employmen s a us is s a is ically
signi ican and posi i e. Hence, ull- ime employed women engage less in gende ed
epo ing, bu o Eas Ge mans, he e is a coun e ac ing e ec . The combined e ec
is no s a is ically signi ican (Coe icien 0.041, SE 0.049, p- alue 0.403), sugges ing
ha o Eas Ge man women he gende -o -in e iewe e ec is no ela ed o he
employmen s a us. Rega ding he ela ion o women’s ela i e income, he esul s
in column 5 imply ha he nega i e e ec o ela i e income on housewo k hou s is
smalle o Eas Ge man women. Fu he mo e, simila o he esul s in Table 6, he
gende -o -in e iewe e ec is lowe when he woman’s con ibu ion o he couple’s
income ises. This nega i e e ec is e e sed o Eas Ge man women, and he com-
bined e ec is again no s a is ically signi ican (Coe icien 0.007, SE 0.173, p- alue
0.967). Hence, o Eas Ge man women he gende -o -in e iewe e ec is also no
signi ican ly ela ed o women’s ela i e income. Gende -o -in e iewe e ec s and
in e ac ion e ms wi h he se e al cha ac e is ics o specializa ion in paid wo k emain
insigni ican o male esponden s (see Tables 25 and 26 in he Appendix).
Summa izing, hese esul s suppo he heo y ha gende -o -in e iewe e ec s
a e less p esen o Eas Ge man women compa ed o Wes Ge man women. G ea e
a achmen o paid wo k gene ally seems o educe gende -o -in e iewe e ec s, bu
o Eas Ge man women his ela ionship is non-exis en . This ep esen s an in e es ing
inding as adi ional social no ms and hence he u ge o ul ill a housewi e ole a e less
p e alen o Eas Ge mans. Addi ionally, hese esul s u he suppo he conclusion
ha he es ablished gende -o -in e iewe e ec s a e indeed d i en by social no ms.
5.5 Gende ed epo ing and cul u al he i age
Se e al s udies ha e shown how social no ms a e ela ed o cul u al he i age (Alesina
e al. 2013; Giuliano 2020). A ecen pape by Blau e al. (2020) s udies he ole o
sou ce coun y cha ac e is ics, especially sou ce coun y gende equali y, in d i ing
he di ision o non-ma ke wo k among US immig an s. The majo esul o he pape
123
A gende gap in gende gaps… 3007
analysis (no shown he e). Impo an ly, he gende -o -in e iewe e ec emains s able
when only including couples who ne e ha e changes in he in e iew mode, which
u he alle ia es conce ns abou he in e iew mode a ec ing he esul s.
7 Conclusion
A g owing body o li e a u e seeks o unde s and how social no ms and social oles
a ec beha io . In his ield o esea ch, he gende ed di ision o housewo k hou s has
ecei ed inc easing a en ion, no only by esea che s bu also in he public deba e.
Ou esul s p esen e idence o he ac ha social no ms impac esponse beha io
in in e iews. We ind a causal e ec o he in e iewe ’s gende on women’s house-
wo k epo ing: Women in couple ela ionships epo mo e ime spen on housewo k
when hey a e in e iewed by a woman ins ead o a man. We do no ind any gende -
o -in e iewe e ec s o male esponden s, which leads o an obse ed in e iewe
gende gap in he housewo k gende gap. The e o e, es ima es abou ime-use om
ace- o- ace in e iews should be in e p e ed wi h cau ion. We a gue ha he obse ed
gende -o -in e iewe e ec is mos likely d i en by sel -decep i e epo ing beha -
io . Women in e nalize socie al expec a ions mo e s ongly when hey in e ac wi h a
emale as opposed o a male in e iewe , which subsequen ly a ec s hei house-
wo k epo . S a i ying he sample and in e ac ing he in e iewe ’s gende wi h
se e al no m- ela ed cha ac e is ics suppo he hypo hesis ha he obse ed gende -
o -in e iewe e ec s a e d i en by social no ms.
No e, while ou esul s p o ide s ong e idence o he exis ence o gende ed epo -
ing in ime-use es ima es, ou indings do no imply ha in e iewe e ec s a e he only
way in which social no ms a ec he epo ing in su eys. By compa ing esponses
owa d emale and male in e iewe s, ou app oach cap u es only a pa o po en ial
adjus men s in he epo ing p ocess. I is possible ha social no ms a ec he epo s
beyond wha is cap u ed by he eac ion o he in e iewe ’s gende . Impo an ly, i
emains an open ques ion whe he e en dia y-based es ima es o ime-use a e in lu-
enced by he social oles indi iduals y o ul ill. As he e a e no da a on he “ ue”
ime-use alues, i is impossible o compa e in e iew epo s o eali y. Inc easing
he use o mixed-me hod app oaches in su ey design could be bene icial o es ing
di e ences be ween a ious modes conce ning ou comes on a wide ange o opics.
While ace- o- ace in e iews a e known o ha e signi ican ad an ages compa ed o
o he in e iew modes (such as building us on sensi i e opics), ou s udy implies
ha d awbacks may exis e en o opics no p e iously conside ed as po en ially
a ec ed by in e iewe e ec s.
Mo e b oadly, ou indings o e a new pe spec i e on he discussion abou no ms
in luencing beha io . While p e ious s udies on socially desi able epo ing ha e
almos exclusi ely ocused on no ma i e iews (e.g., abou gende equali y) o beha -
io s clea ly linked o socie al expec a ions (such as d ug use o o ing beha io ), ou
s udy p o ides e idence ha socially desi able epo ing ma e s e en o opics less
clea ly linked o social no ms. The e o e, ou unde s anding o he impac o no ms
on beha io may be limi ed by he p e alence o epo ing biases o all kinds o ou -
comes. As a consequence, u u e esea ch should ake in o accoun he concei ably
123
3008 M.Ach ,M.Rebaudo
s ong e ec s o epo ing beha io in domains no p e iously ega ded as ele an o
his issue.
Appendix
See Figs. 1,2,3and Tables 11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,
26,27 and 28 in he appendix
Fig. 1 In e iew ques ion - ime-use. Time-use ques ion o he pe sonal in e iew ques ionnai e - SOEP
2018
123
A gende gap in gende gaps… 3009
Fig. 2 Sha e o emale in e iewe s o e ime. The Figu e is based on he main sample as de ined in sec ion
3.The ba s ep esen he sha e o in e iews wi h emale in e iewe s o e ime
Fig. 3 Sha e o emale in e iewe s by in e iew mon h. The Figu e is based on he main sample as de ined
in sec ion 3.The ba s ep esen he sha e o in e iews wi h emale in e iewe s by in e iew mon h. In less
han 4 pe cen o cases in e iews o pa ne s we e in a di e en mon h. Fo simplici y he e only in e iews
wi h emale esponden s a e shown
123
3010 M.Ach ,M.Rebaudo
Table 11 Housewo k ime and in e iewe gende - same day in e iew
Women Men Gap
(1) (2) (3)
Female in e iewe 0.080*** 0.012 0.068***
(0.02) (0.01) (0.03)
Adjus ed R20.165 0.100 0.169
Indi iduals 18149 18149 18149
Obse a ions 97176 97176 97176
The able epo s eg ession analyses o he main sample (cohabi ing couples o age 18 o 65) bu wi hou
couples who we e in e iewed a di e en da es. Dependen a iable is daily housewo k hou s o women,
men o wi hin-couple gap (Housewo k H ou s W oman −Housewo k H ou s Man). The able epo s
he esul s o indi idual ixed e ec s es ima ions. Addi ional con ols as in he main speci ica ion. S anda d
e o s clus e ed a he indi idual le el a e in pa en heses. Signi icance is deno ed by * p<.10, ** p<.05,
*** p<.01
Table 12 Housewo k ime and in e iewe gende - es ic ing housewo k hou s
Housewo k hou s
<=16 <=12 <=8<=6
(1) (2) (3) (4)
Dep. Va : Female housewo k hou s
Female in e iewe 0.079*** 0.077*** 0.063*** 0.056***
(0.02) (0.02) (0.02) (0.02)
Adjus ed R20.170 0.169 0.170 0.161
Indi iduals 18933 18857 17937 16465
Obse a ions 110142 109246 100555 87413
Dep. Va : Male housewo k hou s
Female in e iewe 0.014 0.016 0.018 0.013
(0.01) (0.01) (0.01) (0.01)
Adjus ed R20.097 0.097 0.102 0.099
Indi iduals 18965 18961 18904 18825
Obse a ions 110544 110509 109926 109170
The sample con ains only cohabi ing couples (age 18–65). Con ol a iables as in he main speci ica ion. In
all eg essions we only keep obse a ions who ne e exceed he espec i e h eshold o housewo k hou s.
S anda d e o s clus e ed a he indi idual le el a e in pa en heses. Signi icance is deno ed by * p<.10,
** p<.05, *** p<.01
123
A gende gap in gende gaps… 3011
Table 13 Housewo k ime and in e iewe gende - in e iewe expe ience
Women Men Gap
(1) (2) (3)
Female in e iewe 0.078*** 0.015 0.063***
(0.02) (0.01) (0.02)
In e iewe expe ience 0.002 −0.001 0.003
(0.00) (0.00) (0.00)
Adjus ed R20.167 0.097 0.169
Indi iduals 18965 18965 18965
Obse a ions 110544 110544 110544
The able epo s eg ession analyses o he main sample (cohabi ing couples o age 18 o 65). Dependen
a iable is daily housewo k hou s o women, men o wi hin-couple gap (Housewo k H ou s W oman −
Housewo k Hou s Man). The able epo s he esul s o indi idual (o couple) ixed e ec s es ima ions.
In e iewe Expe ience measu es he cumula i e numbe o yea s in e iewe s ha e wo ked o SOEP.
Addi ional con ols as in he main speci ica ion. S anda d e o s clus e ed a he indi idual le el a e in
pa en heses. Signi icance is deno ed by * p<.10, ** p<.05, *** p<.01
123
3012 M.Ach ,M.Rebaudo
Table 14 Responden cha ac e is ics by in e iewe gende
Mean male in e iewe Mean emale in e iewe Di . SE di .
Income man 2029.94 2087.38 −57.434*** 10.484
Income woman 781.22 767.31 13.904* 5.518
Income HH 3437.45 3461.91 −24.467 12.534
Paid wo k ime man 36.48 36.98 −0.493*** 0.114
Paid wo k ime woman 19.54 19.38 0.157 0.110
Man’s age 45.45 45.18 0.273*** 0.065
Woman’s age 42.70 42.41 0.293*** 0.065
Educa ion le el man 5.01 5.02 −0.011 0.015
Educa ion le el woman 4.94 4.89 0.053*** 0.014
Ma ied 0.88 0.88 0.002 0.002
No. o child en age <60.30 0.30 −0.006 0.004
No. o child en age 6-12 0.43 0.43 −0.002 0.004
No. o child en age 13-16 0.23 0.23 0.003 0.003
In e . day man 95.22 94.75 0.470 0.350
In e . day woman 94.96 94.27 0.693* 0.350
In e iew leng h man 35.10 36.40 −1.300*** 0.081
In e iew leng h woman 34.21 35.39 −1.185*** 0.080
Obse a ions 110,544
The able epo s mean compa isons o esponden s’ cha ac e is ics by in e iewe gende based on he main
sample (cohabi ing couples o age 18 o 65). Paid wo k ime is hou s wo ked pe week including o e ime
hou s. Income is ne mon hly income and household income is ne mon hly income o all household
membe s including non-labo income. Income is co ec ed o in la ion wi h base yea 2016. Educa ion
le els de i e om he CASMIN classi ica ion. Ma ied is a dummy a iable equal o 1 o ma ied couples.
No. o child en is he numbe o child en li ing in he household o each o he age g oups. Signi icance
is deno ed by * p<.10, ** p<.05, *** p<.01
123
A gende gap in gende gaps… 3013
Table 15 Housewo k ime and in e iewe gende - c oss sec ion
Women Men Gap
(1) (2) (3)
Female in e iewe 0.078*** 0.009 0.069***
(0.02) (0.01) (0.02)
Responden Cha ac e is ics
Paid wo k ime woman −0.037*** 0.007*** −0.044***
(0.00) (0.00) (0.00)
Paid wo k ime man 0.008*** −0.018*** 0.026***
(0.00) (0.00) (0.00)
Income woman −0.000*** 0.000*** −0.000***
(0.00) (0.00) (0.00)
Income man −0.000* −0.000*** 0.000**
(0.00) (0.00) (0.00)
Log. HH income −0.010 −0.022 0.012
(0.03) (0.02) (0.03)
Ma ied 0.169*** −0.069*** 0.238***
(0.02) (0.01) (0.02)
No. o child en age <6 0.352*** 0.090*** 0.262***
(0.02) (0.01) (0.02)
No. o child en age 6-12 0.325*** 0.040*** 0.286***
(0.01) (0.01) (0.01)
No. o child en age 13-16 0.325*** 0.026*** 0.299***
(0.01) (0.01) (0.02)
In e iewe cha ac e is ics
Age 0.001 0.000 0.001
(0.00) (0.00) (0.00)
In e iew leng h woman 0.002*** 0.001** 0.001
(0.00) (0.00) (0.00)
In e iew leng h man 0.003*** 0.001*** 0.002***
(0.00) (0.00) (0.00)
Adjus ed R20.415 0.189 0.406
Obse a ions 110544 110544 110544
The able epo s eg ession analyses o he main sample (cohabi ing couples o age 18 o 65). Dependen
a iable is daily housewo k hou s o women, men o wi hin-couple gap (Housewo k H ou s W oman −
Housewo k Hou s Man). The able epo s he esul s in he c oss sec ion wi hou con olling o indi idual
ixed e ec s. Addi ional con ols include age, age squa ed, educa ion ca ego ies o women and men,
espec i ely, in e iew mode, as well as yea and s a e ixed e ec s as in he main speci ica ion. S anda d
e o s clus e ed a he indi idual le el a e in pa en heses. Signi icance is deno ed by * p<.10, ** p<.05,
*** p<.01
123
3014 M.Ach ,M.Rebaudo
Table 16 Housewo k ime and in e iewe gende - exogenous con ols
Women Men Gap
(1) (2) (3)
Female in e iewe 0.082*** 0.010 0.072***
(0.02) (0.01) (0.03)
Responden cha ac e is ics
Ma ied 0.223*** −0.094*** 0.317***
(0.03) (0.02) (0.04)
No. o child en age <6 0.694*** 0.000 0.693***
(0.02) (0.01) (0.02)
No. o child en age 6–12 0.384*** 0.020** 0.364***
(0.02) (0.01) (0.02)
No. o child en age 13–16 0.227*** 0.023*** 0.203***
(0.02) (0.01) (0.02)
In e iewe cha ac e is ics
Age −0.001 0.001 −0.002
(0.00) (0.00) (0.00)
In e iew leng h woman −0.003*** 0.002*** −0.004***
(0.00) (0.00) (0.00)
In e iew leng h man 0.004*** −0.001*** 0.005***
(0.00) (0.00) (0.00)
Adjus ed R20.095 0.024 0.070
Indi iduals 18965 18965 18965
Obse a ions 110544 110544 110544
The able epo s eg ession analyses o he main sample (cohabi ing couples o age 18 o 65). Dependen
a iable is daily housewo k hou s o women, men o wi hin-couple gap (Housewo k H ou s W oman −
Housewo k Hou s Man). The able epo s he esul s o indi idual ixed e ec s es ima ions. Addi ional
con ols include age, age squa ed, educa ion ca ego ies o women and men, espec i ely, in e iew mode,
as well as yea and s a e ixed e ec s as in he main speci ica ion. S anda d e o s clus e ed a he indi idual
le el a e in pa en heses. Signi icance is deno ed by * p<.10, ** p<.05, *** p<.01
123
A gende gap in gende gaps… 3015
Table 17 To al non-ma ke wo k and in e iewe gende
Household To al
cho es non-ma ke wo k
Women Men Women Men
(1) (2) (3) (4)
Female in e iewe 0.058* 0.024 0.208*** 0.040
(0.03) (0.02) (0.08) (0.05)
Adjus ed R20.130 0.146 0.306 0.135
Indi iduals 17549 17548 14459 14462
Obse a ions 98965 98875 72060 72017
The able epo s eg ession analyses o he main sample (cohabi ing couples o age 18 o 65). Dependen
a iable is daily hou s o women and men o he sum o household cho es (housewo k, epai s/ga dening,
e ands) in columns 1 and 2 and o he sum o all non-ma ke ac i i ies (cho es, childca e, elde ly ca e) in
columns 3 and 4. The able epo s he esul s o indi idual ixed e ec s es ima ions. Addi ional con ols
include age, age squa ed, educa ion ca ego ies o women and men espec i ely, in e iew mode, as well as
yea and s a e ixed e ec s. S anda d e o s clus e ed a he indi idual le el a e in pa en heses. Signi icance
is deno ed by * p<.10, ** p<.05, *** p<.01
123
3016 M.Ach ,M.Rebaudo
Table 18 Gende -o -in e iewe e ec s o o he ac i i ies
Childca e Repai /ga den Elde ly ca e E ands Educa ion Sleep Wo k hou s Exe cise Leisu e
(1) (2) (3) (4) (5) (6) (7) (8) (9)
Women
Female in e iewe 0.142*** 0.022* 0.002 0.006 −0.015 −0.006 0.031 0.015 0.053**
(0.05) (0.01) (0.02) (0.01) (0.01) (0.03) (0.05) (0.02) (0.03)
Adjus ed R20.332 0.029 0.004 0.035 0.102 0.017 0.198 0.009 0.066
Indi iduals 18963 18959 14468 17554 18949 9369 18909 7675 18944
Obse a ions 110521 110503 72124 98999 110399 31798 109138 23973 108560
Men
Female in e iewe 0.008 0.009 0.005 0.010 0.006 0.001 0.044 0.005 0.058**
(0.02) (0.01) (0.01) (0.01) (0.01) (0.02) (0.04) (0.02) (0.03)
Adjus ed R20.101 0.057 0.003 0.059 0.099 0.017 0.402 0.014 0.108
Indi iduals 18965 18961 14469 17551 18957 9364 18900 7674 18938
Obse a ions 110531 110413 72128 98993 110374 31797 109304 23971 108524
Dependen a iable is daily hou s o di e en ac i i ies. The sample is es ic ed o couples aged 18 o 65. The able epo s he esul s o indi idual ixed e ec s es ima ions.
Responden s and In e iewe con ols as well as yea and s a e ixed e ec s a e included as in he main speci ica ion. Fo he eg essions wi h wo k hou s as dependen a iable
we exclude ac ual wo k hou s as explana o y a iable (i is no iden ical o he a iable used in he main speci ica ion as he e e.g. a el ime is included bu highly collinea ).
S anda d e o s clus e ed a he indi idual le el a e in pa en heses. Signi icance is deno ed by * p<.10, ** p<.05, *** p<.01
123
A gende gap in gende gaps… 3023
Table 26 Eas in e ac ions - male housewo k hou s
Wes Eas All
(1) (2) (3) (4) (5)
Female in e iewe −0.000 −0.003 0.001 −0.009 −0.024
(0.02) (0.02) (0.02) (0.02) (0.03)
Eas 0.275** 0.281** 0.338***
(0.12) (0.12) (0.13)
Female in e iewe * Eas 0.004 0.027 0.091
(0.03) (0.03) (0.06)
Female in e iewe * woman ull- ime 0.028
(0.02)
Woman ull- ime 0.012
(0.02)
Woman ull- ime * Eas −0.003
(0.03)
Female in e iewe * woman ull- ime * Eas −0.046
(0.04)
Female in e iewe * ela i e income woman 0.083
(0.08)
Rela i e income woman 0.589***
(0.09)
123
3024 M.Ach ,M.Rebaudo
Table 26 con inued
Wes Eas All
(1) (2) (3) (4) (5)
Rela i e income woman * Eas −0.135
(0.11)
Female in e iewe * Eas * ela i e income woman −0.233
(0.15)
Adjus ed R20.037 0.012 0.031 0.031 0.033
Indi iduals 10462 2748 13145 13145 13145
Obse a ions 47792 13865 61657 61657 61657
Dependen a iable is male daily housewo k hou s. The sample is es ic ed o dual-ea ne couples (age 18 o 65). The able epo s he esul s o indi idual ixed e ec s
es ima ions. Responden s and In e iewe con ols as well as yea and s a e ixed e ec s a e included. Full- ime and ela i e income a e s ill de ined o he emale pe son in
he couple o be e in e p e a ion. S anda d e o s clus e ed a he indi idual le el a e in pa en heses. Signi icance is deno ed by * p<.10, ** p<.05, *** p<.01
123
A gende gap in gende gaps… 3025
Table 27 Housewo k ime and global gende gap index (GGI) - yea s spen in Ge many
Women Men
Yea s in Yea s in Yea s in Yea s in
Ge many<20 Ge many>=20 Ge many<20 Ge many>=20
(1) (2) (3) (4) (5) (6)
Female in e iewe 0.122** 0.169*** 0.048 0.030 0.013 0.034
(0.05) (0.06) (0.08) (0.02) (0.03) (0.03)
Female in e iewe * [GGI −GGI]−3.136*** −3.200*** −3.409** −0.133 0.038 -0.020
(1.02) (1.21) (1.58) (0.46) (0.63) (0.64)
[GGI −GGI] 1.201 1.570 0.814 1.352*** 0.818 1.739***
(0.82) (1.02) (1.28) (0.47) (0.61) (0.63)
Yea s in Ge many 0.005 −0.004**
(0.00) (0.00)
Adjus ed R20.350 0.374 0.326 0.165 0.175 0.164
Obse a ions 12813 7298 5515 12826 6095 7130
Reg ession o he subsample o immig an couples in which bo h pa ne s o igina e om he same coun y o o igin (age 18 o 65 in cohabi ing couples). Dependen a iable
is daily housewo k hou s o women, men, o wi hin-couple gap. Couples 1-3 e e o emale esponden s, and columns 4-6 o male esponden s. GGI is he Global Gende
Gap Index. The able epo s he esul s o OLS eg essions. Responden s and In e iewe Con ols as well as yea and s a e ixed e ec s a e included. Addi ional con ols
a e sou ce coun y e ili y and GDP pe capi a. S anda d e o s clus e ed a he indi idual le el a e in pa en heses. Signi icance is deno ed by * p<.10, ** p<.05, ***
p<.01
123
3026 M.Ach ,M.Rebaudo
Table 28 Housewo k ime and in e iewe gende - con ol o in e iew mon h
Women Men Gap
(1) (2) (3)
Female in e iewe 0.080*** 0.011 0.069***
(0.02) (0.01) (0.03)
Feb ua y 0.008 −0.024** 0.032
(0.02) (0.01) (0.02)
Ma ch 0.037 −0.040*** 0.077***
(0.02) (0.01) (0.03)
Ap il 0.043* −0.047*** 0.090***
(0.02) (0.01) (0.03)
May 0.034 −0.030* 0.064**
(0.03) (0.02) (0.03)
June 0.063** −0.027 0.091***
(0.03) (0.02) (0.03)
July 0.082** −0.049*** 0.131***
(0.03) (0.02) (0.04)
Augus 0.069* −0.030 0.098**
(0.04) (0.02) (0.04)
Sep embe 0.143*** −0.020 0.162***
(0.04) (0.02) (0.05)
Oc obe 0.143** −0.059* 0.202***
(0.06) (0.03) (0.07)
No embe 0.081 0.041 0.041
(0.12) (0.06) (0.14)
Decembe 0.207 0.157 0.050
(0.35) (0.11) (0.38)
Adjus ed R20.166 0.100 0.170
Indi iduals 18149 18149 18149
Obse a ions 97176 97176 97176
The able epo s eg ession analyses o he main sample (cohabi ing couples o age 18 o 65) bu wi hou
couples who we e in e iewed a di e en da es. Dependen a iable is daily housewo k hou s o women,
men o wi hin-couple gap (Housewo k H ou s W oman −Housewo k H ou s Man). The able epo s
he esul s o indi idual ixed e ec s es ima ions. Addi ional con ols as in he main speci ica ion. S anda d
e o s clus e ed a he indi idual le el a e in pa en heses. Signi icance is deno ed by * p<.10, ** p<.05,
*** p<.01
Funding Open Access unding enabled and o ganized by P ojek DEAL.
Da a A ailibili y The SOEP da a a e no publicly a ailable bu esea che s can sign a da a dis ibu ion
con ac o access he da a, h ps://doi.o g/10.5684/soep-co e. 35.
The index o gende equali y (GGI) comes om he Wo ld Economic Fo um’s “The Global Gende Gap
Repo ”, a ailable a h ps://www3.we o um.o g/docs/WEF_Gende Gap_Repo _2007.pd .
To al e ili y and GDP da a come om he Wo ld Bank, a ailable a h ps://da a.wo ldbank.o g/indica o /
SP.DYN.TFRT.IN and h ps://da a.wo ldbank.o g/indica o /NY.GDP.PCAP.PP.KD.
123
A gende gap in gende gaps… 3027
Decla a ions
Compe ing in e es s The au ho s ha e no inancial o non- inancial in e es s o disclose ha a e ele an
o he con en o his a icle.
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 Commons 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 ecommons.o g/licenses/by/4.0/.
Re e ences
Ake lo GA, K an on RE (2000) Economics and iden i y. Q J Econ 115(3):715–753
Alesina A, Fuchs-Schündeln N (2007) Good-bye Lenin (o no ?): he e ec o communism on people’s
p e e ences. Am Econ Re 97(4):1507–1528
Alesina A, Giuliano P, Nunn N (2013) On he o igins o gende oles: women and he plough. Q J Econ
128(2):469–530
Baue nschus e S, Raine H (2012) Poli ical egimes and he amily: how sex- ole a i udes con inue o
di e in euni ied Ge many. J Popul Econ 25(1):5–27
Beblo M, Gö ges L (2018) On he na u e o nu u e. The malleabili y o gende di e ences in wo k p e e -
ences. J Econ Beha O gan 151:19–41
Beblo M, Robledo JR (2008) The wage gap and he leisu e gap o double-ea ne couples. J Popul Econ
21(2):281–304
Be and M, Kamenica E, Pan J (2015) Gende iden i y and ela i e income wi hin households. Q J Econ
130(2):571–614
Bi man M, England P, Saye L, Folb e N, Ma heson G (2003) When does gende ump money? Ba gaining
and ime in household wo k. Am J Sociol 109(1):186–214
Blau FD, Kahn LM, Comey M, Eng A, Meye ho e P, Willén A (2020) Cul u e and gende alloca ion o
asks: sou ce coun y cha ac e is ics and he di ision o non-ma ke wo k among US immig an s. Re
Econ Househ 18(4):907–958
Bohlende A, Ra hje M, Glemse A (2020) SOEP-Co e - 2018. Repo o Su ey Me hodology and Field-
wo k
Chapman CD, Benedic C, Schiö h HB (2018) Expe imen e gende and eplicabili y in science. Sci Ad
4(1):e1701427
Col ane S (2000) Resea ch on household labo : modeling and measu ing he social embeddedness o
ou ine amily wo k. J Ma iage Fam 62(4):1208–1233
Cooke LP (2007) Pe sis en policy e ec s on he di ision o domes ic asks in euni ied Ge many. J Ma iage
Fam 69(4):930–950
Deu sches Ins i u ü Wi scha s o schung (2020). Socio-Economic Panel (SOEP), da a o yea s 1984-
2018: e sion 35, SOEP
Fe nández R (2013) Cul u al change as lea ning: he e olu ion o emale labo o ce pa icipa ion o e a
cen u y. Am Econ Re 103(1):472–500
Flo es-Macias F, Lawson C (2008) E ec s o in e iewe gende on su ey esponses: indings om a
household su ey in Mexico. In J Pub Opin Res 20(1):100–110
Fos e G, S a on LS (2018) Do signi ican labo ma ke e en s change who does he cho es? Paid wo k,
housewo k, and powe in mixed-gende Aus alian households. J Popul Econ 31(2):483–519
Geis C (2009) One Ge many, wo wo lds o housewo k? Examining employed single and pa ne ed women
in he decade a e uni ica ion. J Comp Fam S ud 40(3):415–437
Ge shuny J (2013) Na ional u ili y: measu ing he enjoymen o ac i i ies. Eu Sociol Re 29(5):996–1009
Giuliano P (2020) Gende and cul u e. Ox Re Econ Policy 36(4):944–961
123
3028 M.Ach ,M.Rebaudo
Goebel J, G abka MM, Liebig S, K oh M, Rich e D, Sch öde C, Schupp J (2019) The Ge man socio-
economic panel (SOEP). Jah b Na l S a 239(2):345–360
Hausmann R, Tyson L D, Zahidi S (2007) The Global Gende Gap Repo 2007
Hol g a es T (2004) Social desi abili y and sel - epo s: es ing models o socially desi able esponding.
Pe s Soc Psychol Bull 30(2):161–172
Huddy L, Billig J, B acciodie a J, Hoe le L, Moynihan PJ, Pugliani P (1997) The e ec o in e iewe
gende on he su ey esponse. Poli Beha 19(3):197–220
Jessen J, Schweigho e -Kod i sch S, Weinha d F, Be kes J (2024) Sepa a e housewo k sphe es. IZA DP
No. 17134
Kan MY (2008) Measu ing housewo k pa icipa ion: he gap be ween “s ylised” ques ionnai e es ima es
and dia y-based es ima es. Soc Indic Res 86(3):381–400
Kane EW, Macaulay LJ (1993) In e iewe gende and gende a i udes. Pub Opin Q 57(1):1–28
Kan a Public (2019). SOEP-Co e - 2018. Pe sonen agebogen, S ichp oben A-L3+N
K umpal I (2013) De e minan s o social desi abili y bias in sensi i e su eys: a li e a u e e iew. Qual
Quan 47(4):2025–2047
Kühne S (2018) F om s ange s o acquain ances? In e iewe con inui y and socially desi able esponses
in panel su eys. Su Res Me hods 12(2):121–146
Leopold T, Skopek J (2015) Con e gence o con inui y? The gende gap in household labo a e e i emen .
J Ma iage Fam 77(4):819–832
Lippmann Q, Geo gie A, Senik C (2020) Undoing gende wi h ins i u ions: lessons om he Ge man
di ision and euni ica ion. Econ J 130(629):1445–1470
Lipps O, Lu z G (2017) Gende o in e iewe e ec s in a mul i- opic cen alized CATI panel su ey.
Me hods Da a Anal 11(1):67–86
Liu M, S ainback K (2013) In e iewe gende e ec s on su ey esponses o ma iage- ela ed ques ions.
Pub Opin Q 77(2):606–618
Luep ow LB, Mose SL, Pendle on BF (1990) Gende and esponse e ec s in elephone in e iews abou
gende cha ac e is ics. Sex Roles 22(1–2):29–42
Mu ay-Close M, Heggeness M L (2019) Manning up and womaning down: how husbands and wi es epo
ea nings when she ea ns mo e. Oppo uni y and Inclusi e G ow h Ins i u e Wo king Pape s, 28
Paulhus DL (1991) Measu emen and con ol o esponse bias. In: Robinson JP, Sha e PR, W igh sman
LS (eds) Measu es o pe sonali y and social psychological a i udes. Academic P ess, San Diego, pp
17–59
P ess JE, Townsley E (1998) Wi es’ and husbands’ housewo k epo ing: gende class and social desi abili y.
Gend Soc 12(2):188–218
Ro h A, Slo winski M (2020) Gende no ms and income mis epo ing wi hin households. ZEW Discussion
Pape , No. 20-001
Schnell R, K eu e F (2005) Sepa a ing in e iewe and sampling-poin e ec s. J O S a 21(3):389–410
Schobe PS (2013) Ma e nal labo ma ke e u n and domes ic wo k a e childbi h in B i ain and Ge many.
Communi y Wo k Fam 16(3):307–326
Schobe PS, Zoch G (2019) Change in he gende di ision o domes ic wo k a e mo he s o a he s ook
lea e: explo ing al e na i e explana ions. Eu Soc 21(1):158–180
Sulli an O (2010) Changing di e ences by educa ional a ainmen in a he s’ domes ic labou and child
ca e. Sociology 44(4):716–733
Sulli an O (2011) An end o gende display h ough he pe o mance o housewo k? A e iew and eassess-
men o he quan i a i e li e a u e using insigh s om he quali a i e li e a u e. J Fam Theo y Re
3(1):1–13
Sulli an O (2013) Wha do we lea n abou gende by analyzing housewo k sepa a ely om child ca e?
Some conside a ions om ime-use e idence. J Fam Theo y Re 5(2):72–84
Tou angeau R, Rips LJ, Rasinski KA (2000) The psychology o su ey esponse. Camb idge Uni e si y
P ess, Camb idge
Tou angeau R, Yan T (2007) Sensi i e ques ions in su eys. Psychol Bull 133(5):859–883
Wa ne RL (1986) Al e na i e s a egies o measu ing household di ision o labo . J Fam Issues 7(2):179–
195
Wes BT, Blom AG (2017) Explaining in e iewe e ec s: a esea ch syn hesis. J Su S a Me hodol
5(2):175–211
Wes C, Zimme man DH (1987) Doing gende . Gend Soc 1(2):125–151
123
A gende gap in gende gaps… 3029
Zoch G (2021) Gende -o -in e iewe e ec s in sel - epo ed gende ideologies: e idence based on in e -
iewe change in a panel su ey. In J Public Opin Res 33(3):626–636
Publishe ’s No e Sp inge Na u e emains neu al wi h ega d o ju isdic ional claims in published maps
and ins i u ional a ilia ions.
123
