Policy as normative influence? On the relationship between parental leave policy and social norms in gender division of childcare across 48 countries

Abstract

In the present work, we addressed the relationship between parental leave policies and social norms. Using a pre‐registered, cross‐national approach, we examined the relationship between parental leave policies and the perception of social norms for the gender division of childcare. In this study, 19,259 students (11,924 women) from 48 countries indicated the degree to which they believe childcare is (descriptive norm) and should be (prescriptive norm) equally divided among mothers and fathers. Policies were primarily operationalized as the existence of parental leave options in the respective country. The descriptive and prescriptive norms of equal division of childcare were stronger when parental leave was available in a country – also when controlling for potential confounding variables. Moreover, analyses of time since policy change suggested that policy change may initially affect prescriptive norms and then descriptive norms at a later point. However, due to the cross‐sectional nature of the data, drawing causal inferences is difficult.

Full text

Br J Soc Psychol. 2024;00:1–24. wileyonlinelibrar y.com/journal/bjso
|
1
© 2024 Br itish Psycholog ical Societ y.
Receive d: 26 February 2024
|
Accepted: 18 September 2024
DOI: 10. 1111/b jso.1 280 6
ARTICLE
Policy as normative influence? On the relationship
between parental leave policy and social norms in
gender division of childcare across 48 countries
Simon Schindler1 | Carolin Schuster2 | Maria I. T. Olsson3 |
Laura Froehlich4 | Ann- Kathrin Hübner2 | Katharina Block5 |
Colette Van Laar6 | Ton i Sc hm ader7 | Loes Meeussen6,8 |
Sanne van Grootel6 | Alyssa Croft9 | Molly Shuyi Sun10 | Mare Ainsaar11 |
Lianne Aarntzen12 | Magdalena Adamus13 | Joel Anderson14,15 |
Ciara Atkinson9 | Mohamad Avicenna16 | Przemysław Bąbel17 |
Markus Barth18 | Tessa Benson- Greenwald19 | Edona Maloku20 |
Jacques Berent21 | Hilary B. Bergsieker22 | Monica Biernat23 |
Andreea Birneanu24 | Blerta Bodinaku25 | Janine Bosak26 |
Jennifer Bosson27 | Marija Branković28 | Julius Burkauskas29 |
Vladimíra Čavojová13 | Sapna Cheryan30 | Eunsoo Choi31 |
Incheol Choi32 | Carlos C. Contreras- Ibáñez33 | Andrew Coogan34 |
Ivan Danyliuk35 | Ilan Dar- Nimrod36 | Nilanjana Dasgupta37 |
Soledad de Lemus38 | Thierry Devos39 | Marwan Diab40 |
Amanda B. Diekman19 | Maria Efremova41 | Léïla Eisner42 |
Anja Eller43 | Rasa Erentaite44 | Denisa Fedáková13 | Renata Franc45 |
Leire Gartzia46 | Alin Gavreliuc24 | Dana Gavreliuc24 |
Julija Gecaite- Stonciene29 | Adriana L. Germano47 | Ilaria Giovannelli48 |
Renzo Gismondi Diaz49 | Lyudmila Gitikhmayeva50 | Abiy Menkir Gizaw51 |
Biljana Gjoneska52 | Omar Martínez González43 | Roberto González53 |
Isaac David Grijalva54 | Derya Güngör6 | Marie Gustafsson Sendén55 |
William Hall56 | Charles Harb57 | Bushra Hassan58 | Tabea Hässler42 |
Diala R. Hawi59 | Levke Henningsen60 | Annedore Hoppe61 |
Keiko Ishii62 | Ivana Jakšić63 | Alba Jasini6 | Jurgita Jurkevičienė44 |
For affili ations refer to page 19.

2
|
SCHINDLER et al.
Kaltrina Kelmendi64 | Teri A. Kirby65 | Yoko Kitakaji66 |
Natasza Kosakowska- Berezecka67 | Inna Kozytska35 |
Clara Kulich21 | Eva Kundtová- Klocová68 | Filiz Kunuroglu69 |
Christina Lapytskaia Aidy70 | Albert Lee71 | Anna Lindqvist72 |
Wilson López- López73 | Liany Luzvinda16 | Fridanna Maricchiolo74 |
Delphine Martinot75 | Rita Anne McNamara76 | Alyson Meister77 |
Tizita Lemma Melka51 | Narseta Mickuviene29 | María Isabel Miranda- Orrego54 |
Thadeus Mkamwa78 | James Morandini36 | Thomas Morton79 |
David Mrisho78 | Jana Nikitin80 | Sabine Otten81 |
Maria Giuseppina Pacilli82 | Elizabeth Page- Gould10 |
Ana Perandrés- Gómez38 | Jon Pizarro46 | Nada Pop- Jordanova52 |
Joanna Pyrkosz- Pacyna83 | Sameir Quta84 | TamilSelvan Ramis85 |
Nitya Rani86 | Sandrine Redersdorff75 | Isabelle Régner87 |
Emma A. Renström88 | Adrian Rivera- Rodriguez37 |
Rocha- Sánchez Tania Esmeralda43 | Tatiana Ryabichenko41 |
Rim Saab57,89 | Kiriko Sakata66 | Adil Samekin90 | Tracy Sánchez- Pacheco91 |
Carolin Scheifele5,92,93 | Marion K. Schulmeyer49 | Sabine Sczesny94 |
David Sirlopú95 | Vanessa Smith- Castro96 | Kadri Soo11 |
Federica Spaccatini82 | Jennifer R. Steele70 | Melanie C. Steffens93 |
Ines Sucic45 | Joseph Vandello27 | Laura Maria Velásquez- Díaz73 |
Melissa Vink12 | Eva Vives87 | Turuwark Zalalam Warkineh51 |
Iris Žeželj63 | Xiaoxiao Zhang97 | Xian Zhao98 | Yasin Koc81 |
Ömer Erdem Kocak99 | Sarah E. Martiny100
Correspondence
Simon Sc hindler, Federal Un iversity of Appl ied
Admi nist rati ve Sciences, Ber lin, Germany.
Email: dr.simonschindler@g mail.com
Funding information
German Resea rch Foundation, Grant/Award
Number : SCHU 3362/2- 1; Soc ial Sc iences
and Humanit ies Rese arch Counci l Insight
Development Grant, Grant/Award Number:
430- 2 018- 0 0361 and 435- 2014- 1247; Basic
Resear ch Prog ram at the Nation al Rese arch
University H igher School of Economics ( HSE
University); Economic and Social Research
Council, Gr ant/Award Number : ES/S002 74X/1;
Span ish State Research Agency, Grant/Award
Number : PID2019- 111549GB- I00; Gu angdong
13th- five Phi losophy and Socia l Science Plan ning
Project , Grant/Award Number: GD2 0CX L06;
National Natu ral Science Foundation of China,
Abstract
In the present work, we addressed the relationship be-
tween parental leave policies and social norms. Using a
pre- registered, cross- national approach, we examined the
relationship between parental leave policies and the per-
ception of social norms for the gender division of child-
care. In this study, 19,259 students (11,924 women) from
48 countries indicated the degree to which they believe
childcare is (descriptive norm) and should be (prescrip-
tive norm) equally divided among mothers and fathers.
Policies were primarily operationalized as the existence of
parental leave options in the respective country. The de-
scriptive and prescriptive norms of equal division of child-
care were stronger when parental leave was available in a

|
3
PAREN TAL LEAVE POLICY AN D SOCI AL NOR MS
BACKGROUND
The fundamental inf luence and importance of social norms on human behaviour has been extensively
addressed in past interdisciplinary social science research (Legros & Cislaghi, 2020). However, sur-
prisingly little empirical research has been dedicated to understanding how social norms evolve and
change. Changes in norms are important because they represent shifts in individuals' understanding and
interpretation of their society. In the present work, we focus on public policy as one important factor
in shaping social norms. We argue that public policies can signal what is desirable or undesirable within
a society and can influence individuals not only with the force of legal penalties but also by shaping
and promoting social norms. We examined this idea in 48 countries by investigating the relationship
between parental leave policies and young adult's perception of social norms for the gendered division
of childcare among heterosexual couples. In this way, the present study also contributes to the under-
standing of how to close the persistent gender gap in childcare.
The power and evolution of social norms
Social norms inf luence human behaviour powerfully in many aspects of everyday life (Cialdini
et al., 1991). For example, social norms direct us to congratulate people on their birthdays or say thank
you when someone does us a favour; and they proscribe that we do not shout at our supervisors, nor
talk badly about recently deceased people. The importance of social norms as determinants of behav-
iour is outlined in several prominent psychological models, such as the theory of planned behaviour
(Ajzen, 1991) and social role theory (Eagly & Wood, 2012), and a large body of empirical psycho-
logical research demonstrates that social norms are important antecedents of behavioural intentions
(Ajzen, 1991; Armitage & Conner, 2001; Rivis & Sheeran, 2003; Van Kleef et al., 2019).
Due to the popularity of the study of social norms across different research fields, there is variation
in the definition of social norms (Chung & R imal, 2 016; Hogg, 2010; Horne & Mollborn, 2020; Legros
& Cislaghi, 2020). We follow Cialdini and Trost (1998), who define social norms as “rules and standards
that are understood by members of a group, and that guide and/or constrain social behavior without
the force of laws” (p. 152). Accordingly, social norms can communicate what others commonly do (i.e.,
descriptive norms) as well as what others commonly approve or disapprove of (i.e., prescriptive norms; also
called injunctive norms). That is, descriptive norms convey information about what most members of a
group do in given situations, whereas prescriptive norms convey information about how members of a
group should behave in given situations (Goldstein & Cialdini, 2007; Hogg & Reid, 2006).
Grant/Award Number: 31600912; HUM E
Lab Exp erimental Human ities Laboratory,
Facult y of Arts, Masa ryk Un iversity; Swiss
National Science Found ation , Grant/Award
Number : P1ZHP1_184553, P500PS_ 206546
and P2LAP1_194987; Center for Social
Conf lict and Cohesion Stud ies, Grant/Award
Number : 15130009; Center for Interc ultu ral and
Indi genous Researc h, Gra nt/Award Number:
15110006; Social Sc iences and Humanities
Resear ch Council Postdoctora l Fellowship,
Grant/Award Number: 756- 2017- 0249; Slovak
Research and Development Agency, Grant/
Award Numbe r: APVV 20 - 0319; Canad a
Resear ch Cha irs, G rant/Award Numbe r: CRC
152583; Soc ial Sciences a nd Huma nit ies Research
Council, Gr ant/Award Number : 140649; O ntar io
Min istr y of Research and I nnovat ion, Grant/
Award Numb er: 152655
country – also when controlling for potential confounding
variables. Moreover, analyses of time since policy change
suggested that policy change may initially affect prescrip-
tive norms and then descriptive norms at a later point.
However, due to the cross- sectional nature of the data,
drawing causal inferences is difficult.
KEYWORDS
childcare, gender inequality, parental leave, pol icy, social norms

4
|
SCHINDLER et al.
One of the major unresolved questions in the social norm literature refers to how social norms evolve
(Legros & Cislaghi, 2020). Whereas some scholars theorize that behaviour changes first and norms
follow (Morris et al., 2015), others suggest that norms change first and behaviours follow (Mahmoud
et al., 2 014). In our view, assuming mutual influence between these two variables is the most plausi-
ble: the more frequent a behaviour becomes in a certain population, the more individuals will believe
there is a norm, and the more individuals believe in a norm, the more likely they are to comply with it.
However, as other researchers have noted, there is little empirical work on how, precisely, norms evolve
(Bicchieri & Mercier, 2014; Cialdini & Trost, 1998).
Policy as a normative signal
One of the basic mechanisms proposed to underlie norm dynamics involves policymaking (Morris
et al., 2015; Sunstein, 1996). Public policy constitutes a series of attempts made by a government to ad-
dress a public issue by instituting laws and regulations. As such, public policy is presumed to influence
individuals by motivating them to avoid penalties they may incur when law enforcement agents are
present. Beyond punishing undesirable behaviour, legal scholars have proposed that policies have an ex-
pressive function that influences individuals by signalling what is desirable or undesirable within a specific
society: policies express underlying social norms and values and attach a certain normative meaning or
interpretation to a behaviour (McAdams, 2000; Posner, 2000; Sunstein, 1996). Therefore, it has been
claimed that the links between cultural and individual values and norms are mediated through societal
institutions (Schwartz, 2014). In this view, governing institutions are an important source for shaping
social norms (Kinzig et al., 2013; Tankard & Paluck, 2016 ).
Consistent with this view, research has shown that public policies can shape social norms in domains
of smoking bans (Hamilton et al., 2008; Luís & Palma- Oliveira, 2016; Orbell et al., 2009), renewable
energy (Syropoulos et al., 2024), and COVID- 19 lockdowns (Galbiati et al., 2021). Similarly, in two pre-
registered studies – an experimental study (N = 1673) and a longitudinal time- series study (N = 1063)
– Tankard and Paluck (2017 ) found stronger social norms towards support for marriage equality after
a ruling from the U.S. Supreme Court in favour of same- sex marriage. Furthermore, in a natural ex-
periment (N = 437), Eisner et al. (2021), found that informing Swiss participants about a new policy
legalizing stepchild adoption decreased perceived societal disapproval of same- sex parenting compared
with participants not informed about the policy.
In the present work, we moved beyond previous research on policy and social norms by having used
a large sample in an extensive cross- national design. With this design, we were able to isolate how cross-
national variation in policies relates to variation in social norms across different nations, societies, and cul-
tures. To advance theorizing, we explored the idea that public policy decisions can instantly signal what (the
majority thinks) others should do (prescriptive norm) (immediate effect) or change people's behaviour over
time (distal effect), which then in turn changes and signals what people commonly do (descriptive norm).
The case of unequal childcare division
Although both fathers and mothers in Western societies have been spending more time with their chil-
dren in recent decades, fathers' (expected) contributions to the total amount of time parents spend on
childcare among partners in women/man dyads remain rather limited (Dotti Sani, 2020; Dotti Sani &
Treas, 2016; Pailhé et al., 2021; Steinbach & Schulz, 2022; Wei, 2020). Using the same dataset as in the
present work, Olsson et al. (2023), for example, found that across 37 countries, women intended to take
longer leave than men in all countries. Furthermore, fathers' lower engagement in childcare has concern-
ing consequences such as lower career opportunities for women and marital dissatisfaction among couples
(Carlson et al., 2016; Croft et al., 2019), and lower well- being for both fathers and their children (Meeussen
et al., 2020). Existing social norms about gender roles likely play a major role in explaining this gender

|
5
PAREN TAL LEAVE POLICY AN D SOCI AL NOR MS
gap in childcare, as different tasks and behaviours are generally expected of fathers and mothers (Eagly
& Wood, 2012). Traditionally, gender norms favour mothers for childcare tasks: women are expected
to be communal (i.e., caring, warm, social, kind) but not too dominant (i.e., assertive, bossy, arrogant),
while men are expected to be agentic (e.g., competent, independent, rational) but not too weak (e.g., pas-
sive, timid, dependent) (Bosson et al., 2022; Burgess & Borgida, 1999; Croft et al., 2015; De Visser &
McDonnell, 2013; Prentice & Carranza, 2002; Rudman & Fairchild, 2004). To reduce this gender gap in
childcare, it is thus important to understand how social norms upholding a traditional gender division
of childcare can be changed. To do so, we investigate a large selection of countries with varying gender
inequality in childcare and varying parental leave policies. To focus on the relationship between parental
leave policies and norms, we do not examine the actual leave that (new) fathers take (as this is a different
question about whether such policies are effective in incentivizing behaviour) or on people's intentions to
take care of their children (for this, see Olsson et al., 2023). Rather, we examine the relationship between
parental leave policies and young people's estimates of current gender norms. Note that some leave poli-
cies are written exclusively for one parent whereas other leave policies are written to either parent. In the
present work, we refer to parental leave as being available to either parent.
According to the proposed normative power and expressive function of policy, one fruitful strategy
for reducing the gender gap in childcare may involve parental leave policies, as they have been proposed
to not only incentivize actual childcare behaviour, but to also reinforce or change existing gender norms
(Meeussen et al., 2020). Studies on social attitudes support this notion. In a longitudinal study includ-
ing data from nine countries, Omidakhsh et al. (2020) found that changes to parental leave policy that
incentivize or encourage fathers to take time off to care for their children corresponded with changes
in attitudes towards women's equality in the workplace. Furthermore, a study showed that grandparents
whose son had a child after a parental leave reform in Germany in 2007 (including income- dependent
compensation for taking leave, and two of 14 months reserved solely for the father) had more posi-
tive attitudes towards nontraditional gender roles compared to grandparents whose son had a child
shortly before the reform (Unterhofer & Wrohlich, 2 017 ). While personal attitudes are internally mo-
tivated judgements about something (Fishbein & Ajzen, 1975), social norms, instead, are beliefs about
what other people do and approve of. In contrast to the mentioned studies on attitudes (Omidakhsh
et al., 2020; Unterhofer & Wrohlich, 2 017 ), the present research explicitly assesses social norms – that
is, gender norms regarding childcare division.
The present research
In the present work, we argue that public policy can have an expressive function and can influence
behaviours not only with the force of legal penalties or financial incentives, but also by shaping and
promoting social norms. We postulate that the normative influence of public policy decisions can be
immediate by instantly signalling what (the majority thinks) others should do (prescriptive norm) or distal
by changing people's behaviour, which in turn changes and signals what people commonly do (descrip-
tive norm).
Policies vary across nations and cultures. With our large sample of 48 countries, we have a unique
opportunity to examine whether social norms correspond with variations in policies (i.e. a natural ex-
periment). Specifically, we investigated the relationship between parental leave policies on the country
level and the individual perception of social norms regarding childcare division between mothers and
fathers. To get more insights about the potential causality of policy on social norms, we further anal-
ysed the relevance of time since policy decisions were made. This also uncovers potential differences
between descriptive and prescriptive norms (see reasoning below).
We pre- registered the investigation of several country- level predictors that refer to parental leave
policies. First, and most importantly, we assumed that parental leave (i.e., leave that is available to either
parent) constitutes a normative signal that equal childcare division between mothers and fathers is a
socially approved option within society (prescriptive norm) and, over time, leads fathers to engage more

6
|
SCHINDLER et al.
in childcare, leading to perceptions of more equal division of childcare (descriptive norm). Accordingly,
we predicted that prescriptive and descriptive norms of childcare division between mothers and fathers
would be more equal in countries where parental leave is available, compared to countries where it is not
(H1). Note that findings showing leave to be linked to larger gender gaps in intended parental leave uptake
suggest the opposite effect (Boeckmann et al., 2 014; Tharp & Parks- Stamm, 2021).
Second, we assumed that several aspects of the parental leave policy should be beneficial for more
equal childcare division between mothers and fathers. Based on existing evidence that financial gener-
osity of parental leave impacts men more than women (Haas & Hwang, 2 019), we suggested that higher
generosity of parental leave constitutes a normative signal that equal childcare division between mothers
and fathers is a socially approved option within the society (prescriptive norm). This leads especially fa-
thers (compared to mothers) to engage more in childcare and consequently to perceptions of more equal
division of childcare (descriptive norm). Accordingly, we predicted that relatively higher generosity of
parental leave would predict stronger prescriptive and descriptive norms of equal childcare division
between mothers and fathers (H2a).
We further assumed that the extent to which more leave (maternity, paternity, and parental leave) is
exclusively available to mothers (vs. fathers) constitutes a normative signal that mothers are expected
to take the caregiver role (prescriptive norm) and leads mothers to engage more in childcare and con-
sequently to perceptions of more unequal division of childcare at the expense of women (descriptive
norm). Accordingly, we predicted that more exclusive leave for mothers predicts weaker prescriptive and
descriptive norms of equal childcare division between mothers and fathers – with mothers as primary
caregiver (H2b).
Lastly, we addressed the length of parental leave. On the one hand, one could assume that longer
available leave length signals that equal childcare division between mothers and fathers is a socially
approved option within the society (prescriptive norm). On the other hand, the gender gap regarding
intentions to take leave is consistently found to increase with longer possible leave duration (Olsson
et al., 2023; Tharp & Parks- Stamm, 2021). Longer leave length may hence signal desirability of invest-
ment in childcare within traditional gender roles. In light of these contrary predictions, we hypothesized
a bidirectional relationship between prescriptive and descriptive norms regarding gender equality in
childcare division and the available length of parental leave (H2c).
As mentioned above, we postulated that prescriptive norms are instantly affected (immediate effect)
whereas descriptive norms are especially affected over time (distal effect) after a certain policy changed.
To empirically address these assumptions, we explored the effect of time since policy change, that is,
since parental leave was introduced. Accordingly, we reasoned that prescriptive norms regarding gender
equality in childcare division should be perceived as stronger than descriptive norms immediately after
parental leave was introduced as an option. Note that this idea was not pre- registered.
To ensure the robustness of these effects, we controlled for several variables that reflect relevant soci-
etal, national, and cultural differences. These included: gender of participants (individual level), gender
essentialist attitudes (individual level), having children (individual level), and Gross Domestic Product
(GDP) per capita (country level). Especially gender essentialism (i.e., belief that parents' involvement in
childcare is determined by fixed qualities intrinsic to women and men), and GDP per capita (i.e., mea-
sure of a country's economic health) might be confounding variables regarding parental leave policies.
So, we regarded it as informative to check whether the relationship between policies and social norms
holds beyond these variables.
We also pre- registered to control for egalitarianism as a cultural value in each country (i.e., cultural
orientation requiring individuals to see each other as moral equals; Schwartz, 2008). However, including
this variable in the models led to the exclusion of ten countries due to missing values. This means a
substantial reduction of statistical power. Therefore, we decided to report the models including egalitar-
ianism as a control factor only in the Supporting Information on the OSF.
This research was conducted consistent with open science practices. Exclusion criteria, hypotheses,
and analyses were registered prior to data analysis. Pre- registration, materials, data, and procedure are
publicly available on OSF (h t t p s :// o s f . i o / d z q 3 m ).

|
7
PAREN TAL LEAVE POLICY AN D SOCI AL NOR MS
METHOD
Sample
The data used in this paper were collected as part of a large international collaborative research
project aimed at understanding communal orientation in men (https:// ucom2 017. wordp ress.
com/ ). All methods were carried out in accordance with relevant guidelines and regulations.
Collaborators obtained ethical approval from their respective universities (if necessary) and col-
lected data via a questionnaire, either online or in a laboratory. All participants gave informed
consent. Data collection started in October 2017 and ended in June 2019. The data were pre-
pared in accordance with a pre- registered data preparation plan, excluding participants from the
analyses who failed attention checks (e.g. “If you are reading this, please select three”), com-
pleted the questionnaire in less than 10 min, had not been socialized in the specific cultural con-
text before the age of 15 (i.e., moved to their country of residence after age 15), and who did
not fall within the age range 17–30 years. Additionally, individuals from sites that collected
fewer than 6 participants were excluded from analyses as these could not be nested within sites.
The final sample consisted of 19,259 participants (11,924 identified as women, 7078 identified
as men, 257 identified as non- binary; Mage = 20 .54, SD = 2.38) across 123 sites and 48 countries
(see Table 1).
Procedure and instruments
The data were collected using a 45- minute survey which was completed by participants in the language
of instruction at their university. Only relevant items for the present analyses will be described here (for
a complete list, see: https:// osf. io/ rwxcj/ ? view_ only= 35deb 74b4d dc499 58bd7 001a0 064431d).
Outcome variables
Prescr iptive norm
Individual perceptions of prescriptive norms regarding division of childcare were assessed using one
item: “How much of the childcare (taking care of children, spending time with them and fulfilling
their physical and psychological needs) do others in (country) think mothers and/or fathers should do,
respectively?” This was rated on a scale from 0 ( father should do all ) to 100 (mother should do all ); that is, a
value of 50 means equal childcare division.
Descriptive norm
Individual perceptions of descriptive norms regarding childcare division were assessed using one item:
“How much of the childcare (taking care of children, spending time with them and fulfilling their
physical and psychological needs) do mothers and/or fathers do, respectively?”. This was rated on a
scale from 0 ( father does all) to 100 (mother does all ); that is, a value of 50 means equal childcare division.
Policy predictors (country level)
Parental leave availability
We coded whether parental leave available to either parent was available in a country (yes = 1; no = 0).
Data were obtained from the International Labour Organization (2014) report. Of the 48 countries,
parental leave was available in 30 countries (62.5%; n = 13,483; see Table 1).

8
|
SCHINDLER et al.
Financial generosity of parental leave
An index was computed to capture the generosity of leave available to both mothers and fathers in each
country (see Olsson et al., 2023). This index is the duration of parental leave (in weeks) multiplied by
the rate of compensation (percentage of earnings prior to leave). The resulting indicator represents the
number of weeks of 100 percent income (e.g., 10 weeks compensated at 80% would be 8 weeks). Data
were obtained from the International Labour Organization (2014) report (range: 0–78 weeks). Values
were grand mean centred.
Gender imbalance in exclusive leave
Following the procedure of Olsson et al. (2023), an index was computed to capture the ratio of how
much of the parental leave is exclusive to mothers relative to fathers in each country. The index is calcu-
lated as the duration of maternity leave (in days) + duration of parental leave (in days) that are exclusively
reserved for mothers – the duration of paternity leave (in days) – duration of parental leave (in days) that
are exclusively reserved for fathers. Positive scores indicate more unequal leave policies (in favour of the
mother). Data were obtained from the International Labour Organization (2014) report (range: −10 to
283 days) and grand mean centred.
Available parental leave length
Available leave represents the total amount of leave (in weeks) that is available to either parent (i.e., no
part of this leave is exclusive to mothers or fathers). Data were obtained from the International Labour
Organization (2014) report (range: 0 –156 weeks) and was grand mean centred.
TABL E 1 Sample size by country and informat ion about whether parental leave is available.
Countr y nParental leave Countr y nParental leave
Albania 154 Ye s Lit huan ia 194 Yes
Australia 450 Ye s Nort h Macadon ia 159 Yes
Belgiu m 385 Yes Malaysia 342 No
Bolivia 339 No Mexico 199 No
Canada 1333 Yes Netherlands 554 Ye s
Chile 447 Yes New Zealand 242 Yes
China 169 No Nor way 305 Ye s
Colombia 411 No Pa kistan 215 No
Costa Rica 219 No Palestine 121 No
Croatia 424 Yes Poland 515 Yes
Czech Republic 217 Yes Romania 237 Yes
Denmark 157 Ye s Russia 187 Yes
Ecuador 185 No Serbia 778 No
Estonia 213 Ye s Singapore 210 No
Ethiopia 203 No Slovak ia 277 Ye s
France 429 Ye s Sout h Korea 157 Yes
Germany 681 Ye s Spain 381 Yes
India 152 No Sweden 198 Ye s
Indonesia 2 51 No Switzerland 1092 No
Ireland 304 Ye s Tanzania 120 No
Italy 300 Yes Tu rk e y 580 No
Japan 512 Yes Ukraine 315 Yes
Kazak hstan 156 Yes U.K. 285 Yes
Lebanon 19 0 No U.S .A. 3315 Yes

|
9
PAREN TAL LEAVE POLICY AN D SOCI AL NOR MS
Introduction of parental leave availability
According to the available data from the International Labour Organization (2014), we coded countries
with no parental leave available, countries that had parental leave available since 2013, and countries
that had parental leave since 1994. Only data for these dates were reported in the International Labour
Organization (2014).
Control predictors (for robustness checks only)
Gender of participants
Gender was measured by the item: “What best reflects your gender?”. Possible responses: “male” (coded
as 1), “female” (coded as 0) and “neither best ref lects my identity”.
Gender essentialist attitudes
Gender essentialist attitudes were assessed on the individual level with three items proposed by
Gaunt (2006): “Mothers are instinctively better caretakers than fathers”, “Mothers are naturally more
sensitive to a baby's feelings than fathers are”, and “In terms of childcare, fathers have to learn what
mothers are able to do naturally”. Response options ranged from 1 (strongly disagree) to 7 (strongly agree). In
an internal consistency calculation of the three items, all countries showed a Cronbach's Alpha of ≥.70,
so a composite score of the three items was computed (as specified in the pre- registration for data clean-
ing). Only one country (Ethiopia) achieved a value <.55 and was therefore excluded from robustness
checks. Values were grand mean centred (Enders & Tofighi, 2007).
Having children
Whether participants have children was assessed using the item: “In your future, do you expect you
will have children?”, where the response “I already have a child/children” was coded as 1 and all other
answers were coded as 0.
Gross domestic product (GDP) per capita
Although not pre- registered, we decided to additionally control for countries' GDP (per capita) as a
comprehensive measure of economic performance since this variable has been found to be linked
to sociopolitical developments (Korotayev et al., 2018). We opted to use GDP values from 2017, be-
cause our data collection started in 2017. Values ranged from about $US 767 (Ethiopia) to 80.189 $US
(Switzerland) per capita.
Data analyses
Data analyses followed the pre- registered protocol (except where noted). We first checked the normal-
ity distribution and skewness of the predictor variables. Financial generosity (skewness = 2.22), gender
imbalance in exclusive leave (skewness = 1.52), and available leave length (skewness = 1.19) were right-
skewed and non- normally distributed. Therefore, we used Spearman's rank- order correlations when
calculating bivariate correlations between the predictor variables and the perceived descriptive and pre-
scriptive norm. For robustness checks, we further recoded these three continuous variables into ordinal
variables and re- ran the regression models. We ran linear- mixed models (LMM ) by nesting individual
data within sites and within countries. Random effects for sites and countries were included in the
models. Separate models were used for the prescriptive and the descriptive norm measures. The control
predictors were only included for the robustness checks (not the main analyses). All predictor variables
were entered as fixed factors. Due to a likely confound between the leave availability and the other three
main predictors, we ran separate models (LMM 1: leave availability as main predictor; LMM 2: financial
generosity, gender imbalance in exclusive leave, and available leave length as continuous predictors).

10
|
SCHINDLER et al.
RESULTS
The means of the prescriptive and descriptive norms regarding the gender division of childcare (see
Table 2) were significantly larger than the equality value of 50, both ps < .001, both C ohen's ds > 12.29,
clearly indicating that across all countries, participants estimate the existence of gender norms dictating
that mothers should and are perceived to actually do more childcare than fathers (means across coun-
tries ranged from 57.10 to 83.00 for the prescriptive norm and 59.30 to 87.80 for the descriptive norm).
Although the correlation between the two norms (see Table 2) was positive, significant, and strong
(Spearman's ρ = .60; Cohen, 198 8), the size of this effect suggests that the two norms are distinct.
Hypothesis testing
Bivariate correlations
Due to the right- skewed distribution of financial generosity, gender imbalance, and available leave, we
calculated Spearman's rank- order correlations for all variables. Bivariate correlations can be found in
Table 2. Most importantly, and as expected, there were negative correlations between both types of
norms and the availability of leave, indicating that in countries that have policies providing parental
leave (i.e., leave that both mothers and fathers can take) gender equality in childcare division is more
promoted as a social norm. The correlation was large for the prescriptive norm (who should take pa-
rental leave, Spearman's ρ = −.53), whereas the correlation for the descriptive norm was moderate (who
does take parental leave, Spearman's ρ = −.32). Significant negative correlations between prescriptive
norms and leave policies (i.e., financial generosity and amount of available leave length) further mean
that higher financial generosity and higher amount of available leave length are positively associated
with the belief that gender division in childcare should be equal. For descriptive norms, only the cor-
relation with leave availability was significant.
LMM 1: Parental leave availability
We first tested the hypothesis that prescriptive and descriptive norms of childcare division between
mothers and fathers would be more equal in countries where parental leave is available, compared
to in countries where it is not (H1). In line with this hypothesis, the two LMMs using the policy
variable parental leave availability as a binary predictor significantly predicted the prescriptive norm
TABL E 2 Mea ns, sta ndard deviat ions, and intercorrelations (Spearman's ρ) of study variables (N = 48 cou ntries).
Mean SD (1) (2) (3) (4) (5)
(1) Prescript ive norm 67.42 5.48 –
(2) Descriptive norm 68.25 4.82 .60 –
(3) Leave ava ilability – – −.53 −.32 –
(4) Financial generosit y of leave 9.90 17. 42 −. 34 −.11 .60 –
(5) Gender imbalance in exclusive leave 94.13 68.03 .09 .10 .36 .07 –
(6) Available leave length 51.48 62.88 −.40 −.13 .87 .59 .33
Note: Bold m arked correlations a re signif icant at p < .05. Indi vidua l scores of prescript ive and descriptive nor ms were averaged for e ach country.
Means a nd standard dev iations refer to t he count ry leve l. The scales of t he norm va riable s range from 0 (men shou ld do/do all the c hildcare) to
100 (women shou ld do/do all the childcare); a value of 50 means equal childcare d ivision. Leave availabil ity was coded wit h ‘yes’ as 1 and ‘no’
as 0. The scale of financial g enerosity ind icates the numbe r of weeks of 100 percent income. The scale of gender imbala nce in exc lusive leave
indicates the ratio of how much of the pa renta l leave is exclusive to mothers relati ve to fathers in ea ch count ry. Posit ive scores i ndicate more
unequal leave policie s in favou r of the mot her. Avai lable leave length represents the total a mount of leave in weeks that is avail able to eit her
parent.

|
11
PAREN TAL LEAVE POLICY AN D SOCI AL NOR MS
and the descriptive norm (see Table 3). Thus, in countries where parental leave is available, the norm
that women should do (prescriptive norm) and actually do (descriptive norm) all the childcare was
weaker and close to a more equal division. The prediction was descriptively stronger for prescriptive
norms and was still significant with p < .001 for prescriptive norms when the descriptive norm was
included in the model.
Adding gender, having children, essentialism, and GDP per capita as control variables to the model
(LMM 1_r; see Table 4), parental leave availability as a predictor was still significant for the prescriptive
norm ( p = .012) but not significant for the descriptive norm ( p = .330). Notably, each control variable
(except having children in the model for the prescriptive norm) was a significant predictor in these ro-
bustness analyses (see Discussion).
LMM 2: Financial generosity, gender imbalance in exclusive leave and leave length
The LMMs using the policy variables financial generosity (H2a), gender imbalance in exclusive leave
(H2b), and leave length (H2c) as predictors yielded no significant predictions for the prescriptive norm
nor the descriptive norm (see Table 5). When including gender, having children, essentialism, and GDP
as control variables the models also did not predict the prescriptive ( ps > .107) or descriptive norm
(ps > .509). As a robustness check, we recoded the three continuous predictors into ordinally scaled
variables, with no change in the significance of the models (prescriptive norm: ps > .116; descr iptive
norm: ps > .484). More information on the recoding procedure and detailed results can be found in the
Supporting Information on the OSF.
Since some of the policy variables were correlated, we ran separate LMMs for each of the three policy
predictors. In line with H2a, financial generosity predicted prescriptive norms, b = −0 .0 9, SE b = 0.04 ,
p = .048. No significant effects occurred for gender imbalance in exclusive leave ( p = .162) and leave
length ( p = .163). There were no significant predictions for the descriptive norm (all three ps > .386 ).
Note that these separate analyses were not pre- registered.
TABL E 3 Fixed effects and random effects in the L MM 1 for prescriptive and descriptive norm.
LMM 1 for prescriptive norm
(n = 19,23 0; 48 cou ntries) b SE b
95% CI
pLL UL
Fixed effects
Intercept 71.14 1.07 69.02 73.28 <.0 01
Parenta l leave availabi lit y −5.98 1.34 −8.68 −3. 31 <.001
Random effects b SD b
Intercept variance (site- level) 2.67 1.64
Intercept variance (country- level) 27.42 4 .17
LMM 1 for descriptive norm
(n = 19,22 9; 48 count ries)
Fixed effects
Intercept 70.54 1.03 68.49 72.60 <.001
Parenta l leave availabi lit y −3.72 1.30 −6.32 −1.13 .006
Random effects b SD b
Intercept variance (site- level) 1.87 1.37
Intercept variance (country- level) 16 .62 4.08
Note: Bold m arked correlations a re signif icant at p < .05. Parental leave avail abil ity was coded wit h ‘yes’ as 1 and ‘no’ as 0.

12
|
SCHINDLER et al.
Time passed since introduction of parental leave
To explore immediate and distal normative effects of the introduction of parental leave, we considered
the time passed since the introduction of parental leave. We categorized countries into three groups
according to the data from the International Labour Organization report: no parental leave policy in
place, available since 2013, and available since 1994. We used dummy- coded variables to predict child-
care division norms. Results can be found in Table 6 and Figure 1.
The norm that women should do (prescriptive norm) and actually do (descriptive norm) all childcare
was significantly weaker when parental leave had been available since 1994 (vs. no policy in place). When
parental leave was only made available in 2013, only the prescriptive norm was significantly weaker
compared to when no parental leave policy was in place. The effect for the descriptive norm was not sig-
nificant. That is, the effect on the descriptive norm – in contrast to the prescriptive norm – was only sig-
nificant when parental leave had been introduced a longer time ago. Including gender, having children,
essentialism, and GDP as control variables yielded a significant comparison for the prescriptive norm
(p = .004): the norm that women should do all childcare was significantly weaker when parental leave
was available since 1994 (vs. no policy in place). The comparison between introduction of leave in 2013
and ‘no policy in place’ approached significance for the prescriptive norm ( p = .056). No si gnif icant
TABL E 4 Robustness analyses for the fixed effects and random effects in t he LM M1_r for prescript ive and descript ive
norm.
LMM 1_r for prescriptive norm
(n = 18,722 , 47 countr ies) b SE b
95% CI
pLL UL
Fixed effects
Intercept 69.09 0.94 6 7.19 70.98 <.001
Parenta l leave availabi lit y −2.88 1. 09 −5.07 −0.69 .012
Gender −3.40 0.25 −3.88 −2.92 <.001
Having children 2.12 1 .11 −0.06 4.29 .056
Gender essentialism 0.24 0.08 0.07 0.40 .004
GDP per capita −0.13 0.02 − 0 .17 −0.08 <.001
Random effects b SD b
Intercept variance (site- level) 2.30 1.52
Intercept variance (country- level) 8.46 2.91
LMM 1_r for descriptive norm
(n = 18,720 ; 47 countr ies)
Fixed effects
Intercept 69.00 0.95 6 7.11 70.90 <.001
Parenta l leave availabi lit y −1.08 1.10 −3.27 1.11 .330
Gender −3.21 0.24 −3.6 8 −2.73 <.0 01
Having children 3.55 1.10 1.40 5.70 .001
Gender essentialism 1.16 0.08 1.0 0 1.33 <.001
GDP per capita −0.06 0.02 − 0.11 −0.01 . 017
Random effects b SD b
Intercept variance (site- level) 1.89 1.37
Intercept variance (country- level) 8.80 15.53
Note: Bold m arked correlations a re signif icant at p < .05. Parental leave avail abil ity was coded wit h ‘yes’ as 1 and ‘no’ as 0. Gender was coded
with ‘m ale’ as 1 a nd ‘fema le’ as 0. Having children wa s coded with ‘yes’ as 1 and all othe r options as 0. Higher score s in gender essent ialism
indicate a stronger gender esse ntia list at tit ude. Or igi nal va lues for GDP pe r capita ( in $US) were div ided by 1.000 to increase re adabi lit y of the
corresp onding coeff icient estim ates. Et hiopia was excluded as a country due to low reliabil ity in gender essenti alism (Cronbach's alpha < .55 ).

|
13
PAREN TAL LEAVE POLICY AN D SOCI AL NOR MS
TABL E 6 Fixed effects and random effects in the L MM for prescriptive and descriptive norm with i ntroduction of
parental leave (dum my coded) as predictor.
LMM for prescriptive norm
(n = 19,23 0; 48 cou ntries) b SE b
95% CI
pLL UL
Fixed effects
Intercept 71.13 1.02 69.1 73.18 <.001
Available since 2013 (dummy 1) −4.52 1.46 −7.4 4 −1.62 .003
Available since 1994 (dummy 2) −7.8 3 1.56 −10.97 −4.73 <.001
Random effects b SD b
Intercept variance (site- level) 2.66 1. 63
Intercept variance (country- level) 15.68 15.86
LMM for descriptive norm
(n = 19,22 9; 48 count ries)
Fixed effects
Intercept 70.54 1.00 68.53 72 .55 <.001
Available since 2013 (dummy 1) −2.67 1.43 −5.54 0.20 .070
Available since 1994 (dummy 2) −5.06 1.54 −8.14 −1.99 .002
Random effects b SD b
Intercept variance (site- level) 1.84 1.36
Intercept variance (country- level) 15.7 3.96
TABL E 5 Fixed effects and random effects in the L MM 2 for prescriptive and descriptive norm.
LMM 2 for prescriptive norm
(n = 19,23 0, 48 cou ntries) b SE b
95% CI
pLL UL
Fixed effects
Intercept 67.38 0.72 65.94 68.83 <.001
Financial generosity −0.06 0.05 −0.16 0.04 .228
Gender imbalance in exclusive leave 0.02 0.01 −0.01 0.04 .195
Available leave length −0.01 0.01 −0.04 0.02 .409
Random effects b SD b
Intercept variance (site- level) 2 .70 1.64
Intercept variance (country- level) 22.31 4.72
LMM 2 for descriptive norm
(n = 19,22 9, 4 8 countr ies)
Fixed effects
Intercept 68.20 0.67 <.001
Financial generosity −0.03 0.05 .584
Gender imbalance in exclusive leave 0.00 0.01 .787
Available leave length −0.00 0.01 .720
Random effects b SD b
Intercept variance (site- level) 1.86 1.37
Intercept variance (country- level) 19.42 4.41
Note: Bold m arked correlations a re signif icant at p < .05. Fina ncia l generosity and available time of leave we re assessed in week s. Gende r
imbal ance i n exclusive leave was assessed i n days. A ll predictors were grand mea n centred.

14
|
SCHINDLER et al.
predictions occurred for the descriptive norm when including the control variables ( ps > .210). Detai led
results can be found in the Supporting Information on the OSF.
DISCUSSION
Despite extensive research on the influence of social norms on human behaviour (Legros &
Cislaghi, 2020), little empirical work has been conducted on how these norms evolve and change
(Bicchieri & Mercier, 2014). In the present work, we addressed the potential expressive function of
policymaking for shaping social norms. Although this idea is widespread in the literature, it has received
little attention on an empirical level (for exceptions, see e.g., Eisner et al., 2021; Tankard & Paluck, 2 017 ).
Using a pre- registered, large- sample, cross- national approach, we addressed this gap. Specifically, we
examined the relationship between policy at a national level and individuals' perceptions of prescriptive
and descriptive norms in the context of the division of childcare between mothers and fathers. With this
study design, we were able to examine the proposed relationship across different cultural, societal, and
political systems. We were also able to examine a time- relevant aspect of this – that is, whether the rela-
tionship between polices and prescriptive or descriptive norms, respectively, differs between countries
where policies were adopted earlier compared to countries where policies were adopted later.
The relationship between policy and norms
First, in line with previous research, we found generally prescriptive and descriptive norms in favour
of mothers doing more childcare than fathers. Importantly, however, as predicted (H1), results of the
bivariate correlations and the corresponding LMM showed that in countries where parental leave is
available, the norm that women should do (prescriptive norm) and actually do (descriptive norm) more
FIGURE 1 Effects of introduction of parental leave on prescript ive and descript ive norms regard ing gender division in
childcare.
50
55
60
65
70
75
80
85
90
95
100
Prescriptive normDescriptive norm
Normsregarding gender division of childcare
No leave policy
Since 2013
Since 1994
Mothers
(should) do all
Equal
divisi on
p < .001
p = .003
p = .002
p = .070

|
15
PAREN TAL LEAVE POLICY AN D SOCI AL NOR MS
childcare was significantly weaker, indicating a stronger tendency towards a more equal division. For
the prescriptive norm, this relationship was weakened, but still significant in the LMM, when control-
ling for gender, having children, essentialism, and GDP per capita. This supports the notion that policy
making in this context plays a unique role – at least for prescriptive norms. Interestingly, each of the
control variables (except having children in the model for the prescriptive norm) was a significant pre-
dictor in these robustness analyses indicating stronger prescriptive and descriptive norms of equal divi-
sion in childcare for (a) male (vs. female) participants, (b) for participants having no children (vs. having
children; only for the descriptive norm), (c) for individuals with weaker gender essentialist attitudes and
(d) for countries with higher GDP per capita. Beyond the included control variables, there are further
institutional (country- level) variables that might explain the link between the availability of parental
leave and norms of gender division in childcare, such as expansion of public childcare, female labor
force participation or policy landscape. Future research should investigate these possible confounds.
Results of the bivariate correlations provided further support for our hypotheses: financial generos-
ity of leave (H2a) and the available leave length (H2c) were negatively related to individuals' perceptions
of the prescriptive norm. That is, higher financial generosity of leave and longer availability of leave are
associated with prescriptive norms favouring a more equal division of childcare. There was no signifi-
cant correlation with gender imbalance in exclusive leave. Furthermore, financial generosity, available
leave length, and gender imbalance in exclusive leave did not predict prescriptive or descriptive norms
in the LMMs, when being included simultaneously. Financial generosity only predicted prescriptive
norms when included as a single predictor. Thus, regarding the prescriptive norm, H2a was supported
by this result and by the bivariate correlation, while H2c was only supported by results of the bivariate
correlation. Taking the nested data structure into account, yielded no strong support for the inf luence
of these variables. These null findings could indicate that these specific parental leave policies do not
have a substantial impact on social norms. It could be, for example, that people are not aware of these
policies and/or that they only weakly signal gender equality in childcare division (see Eisner et al., 2021).
However, present null f indings might also be due to low statistical power in some of the analyses (see
Limitations below).
Time since policy change
Exploring the role of time since new leave policies were introduced (distal vs. immediate), revealed that
the norms that women should (prescriptive norm) and actually do (descriptive norm) all childcare were
both significantly weaker when parental leave had been available since 1994 (distal) compared to when
parental leave was not available. When parental leave was available since 2013 (immediate), the prescrip-
tive norm was significantly weaker compared to when parental leave was not available, suggesting an
immediate effect on prescriptive norms whereas the effect for the descriptive norm was not significant.
These results were basically not affected when taking gender, having children, essentialism, and GDP
per capita into account as control variables. Taken together, these findings suggest that policymaking
may more quickly affect perceptions of prescriptive norms, whereas changes in perceived descriptive
norms take more time. In other words: it seems like people first interpret a new policy as a normative
signal about what should be done, and it takes more time until they see this reflected in what other people
in their country actually do. Over time, policies appear to be linked to both types of norms.
Descriptive versus prescriptive norms
In sum, our results provide support for the expressive function of policymaking for shaping social
norms. Interestingly, relationships were consistently larger and more robust for the prescriptive norm
than the descriptive norm, speaking for a more immediate normative influence of public policy on pre-
scriptive (what should be done) than descriptive norms (what is done). The present findings favour the

16
|
SCHINDLER et al.
prescriptive norm as the first norm to change rather than the descriptive norm. However, this might
be restricted to the context of social issues, or childcare division and parental leave policies specifically.
In addition, many scholars argue that a social norm could not become prescriptive at all unless it was
descriptive initially (Morris et al., 2015). In our view, at least in the context of policymaking, assuming
mutual influence between the two norms remains most plausible (Eriksson et al., 2015) – especially over
time (Luís & Palma- Oliveira, 2016; Nyborg, 2003).
Practical relevance
Besides contributing to the theoretical understanding of how social norms evolve, our findings also
point to the potential power of policies for shaping and promoting behaviour, beyond the force of legal
penalties. These changes in perceived social norms matter because they represent shifts in individuals'
understanding of their society—where it stands and where it is going. Nevertheless, for the efficient
application of the expressive function of policymaking, it is important to investigate the boundary
conditions for the effect of policies. For example, one could assume that some level of trust in govern-
ment is necessary for the expressive function of policies and that political orientation is likely to play a
crucial role (Tankard & Paluck, 2017 ). Furthermore, the effect of policies on social norms might further
depend on whether current policies are salient or not, or whether and how they were communicated in
the public media.
Limitations
The present cross- sectional data do not allow causal inferences. Although we theoretically addressed
the question of whether policymaking impacts social norm shifts, it is also plausible that existing social
norms impact policymaking. For example, public opinion influences policies through political voting
decisions. Nevertheless, the existence of a relationship between policy and social norms is a necessary
condition for a causal effect of policy, and the present research thus makes an important step in showing
this relationship. Additionally, the documented relevance of time passed since policy decisions points to
a causal effect of policy on social norms. While applying a pre- post design would provide stronger sup-
port for causality (Tankard & Paluck, 2 017 ), our data allow generalizing the relationship between policy
and social norms across a large selection of countries.
On the individual level, our sample includes more than 19,000 participants. However, with only 48
cases on the country level, statistical power in the analyses is rather low (meaning a high probability of
false negative errors). Thus, especially, the present null findings should be interpreted with caution. For
this reason, we refrained from further (theoretically potentially insightful) exploratory subgroup analy-
ses. One could, for example, conduct separate analyses for geographically or culturally close countries,
such as European Welfare States. This exemplary subgroup would consist of 16 countries, thus, a sub-
stantially smaller sample reducing statistical power even more. Especially, null findings would be highly
fragile and speculative under these circumstances. To control for cultural similarity, we pre- registered
analysis including egalitarianism as a cultural value in each country (Schwartz, 2008). However, as al-
ready mentioned above, including this variable in the models led to the exclusion of ten countries due
to missing values. This led to a substantial reduction of statistical power. That is, changes in the models
through including egalitarianism might occur (a) because of the missing values for egalitarianism in ten
countries or (b) because controlling for egalitarianism might have removed real shared variance because
egalitarianism and policies are affected by similar variables and affect one another. So, it remains open
question whether including egalitarianism or not reveals the more accurate result.
The present work does not address actual behaviour or behavioural intentions. However, parental
leave policies were previously shown to have an impact here (Olsson et al., 2023). For example, longitu-
dinal studies have shown that introducing incentives for fathers to take parental leave increases uptake

|
17
PAREN TAL LEAVE POLICY AN D SOCI AL NOR MS
in men ( Jurado- Guerrero & Muñoz- Comet, 2021). Future research should thus investigate the mediat-
ing role of social norms for policy effects on actual behaviour.
Generalizability is limited by the used sample of relatively young, educated participants, who likely
are anticipating but not yet directly involved with the issues of childcare division and parental leave
policies. We tried to address this issue by controlling for whether participants had one child or more
children. This did not change our results. Besides that, investigating university students is an informa-
tive endeavour, as it allows us to better understand the decisions they will make regarding work- family
divisions in the future. Moreover, young, highly educated individuals are more likely to later hold posi-
tions of power and influence policies at an organizational or national level ( Meeussen et al., 2016 , 2019).
By gaining insight into the choices they make now, we can better understand the decisions they might
make in the future when they are in a position to influence the lives of others. As a result, the perceived
social norms of this group may provide insight into the development of societies.
Research on the link between policy and anti- gay (or anti- queer) attitudes debates a backlash effect in
terms of greater disapproval of the issue induced by policy change – however, the evidence for a backlash
effect is weak (Bishin et al., 2016; Flores & Barclay, 2016). This possibility should, however, be taken
into account when investigating the effect of policy on social norms. There is also the argument that
once something becomes policy, people are motivated to justify it as part of the system (Laurin, 2 018).
Beyond the expressive function of policy, this could also explain why norms of equal gender division
in childcare are stronger in countries where policy enables parental leave. Further research is needed to
investigate the mechanisms of potential policy effects more thoroughly.
CONCLUSION
To our knowledge, the present work is the first large- scale cross- national approach investigating the
relationship between policy and social norms. Assessing the prescriptive and descriptive norm regard-
ing childcare division between mothers and fathers in 48 countries, we found support for a relationship
between parental leave policy and these norms, indicating that introducing parental leave availability
may reduce the norm that mothers should (prescriptive norm) and actually do (descriptive norm) all
the childcare. Moreover, analyses of time since policy change suggested that policy change may initially
first affect prescriptive norms and then descriptive norms at a later point. In sum, our findings provide
(partial) empirical support for the expressive function of policy. Nevertheless, due to the cross- sectional
nature of the data, the present results should be interpreted with caution and should not be understood
as evidence for causal mechanisms.
AUTHOR CONTRIBUTIONS
Simon Schindler: Conceptualization; methodology; data curation; writing – original draft; writing –
review and editing; formal analysis. Carolin Schuster: Conceptualization; funding acquisition; writing
– review and editing; writing – original draft; methodology. Maria I. T. Olsson: Writing – review and
editing; writing – original draft. Laura Froehlich: Writing – original draft; writing – review and edit-
ing. Ann- Kathrin Hübner: Writing – original draft; writing – review and editing. Katharina Block:
Writing – original draft; writing – review and editing. Colette Van Laar: Writing – original draft; writ-
ing – review and editing. Toni Sc h mader: Writing – original draft; writing – review and editing; fund-
ing acquisition. Loes Meeussen: Writing – review and editing; writing – original draft. Sanne van
Grootel: Writing – original draft; writing – review and editing. Alyssa Croft: Writing – review and edit-
ing; writing – original draft. Molly Shuyi Sun: Writing – original draft; writing – review and editing.
Mare Ainsaar: Writing – original draft; writing – review and editing. Lianne Aarntzen: Writing – re-
view and editing; writing – original draft. Magdalena Adamus: Writing – original draft; writing – re-
view and editi ng. Joel Anderson: Writing – original draft; writi ng – review and edit ing. Ciara Atkinson:
Writing – original draft; writing – review and editing. Mohamad Avicenna: Writing – original draft;
writing – review and editing. Przemysław Bąbel: Writing – original draft; writing – review and editing.

18
|
SCHINDLER et al.
Markus Barth: Writing – original draft; writing – review and editing. Tessa Benson- Greenwald:
Writing – original draft; writing – review and editing. Edona Maloku: Writing – original draft; writing
– review and editing. Jacques Berent: Writing – original draft; writing – review and editing. Hilary B.
Bergsieker: Writing – original draft; writing – review and editing. Monica Biernat: Writing – original
draft; writing – review and editing. Andreea Birneanu: Writing – original draft; writing – review and
editing. Blerta Bodinaku: Writing – original draft; writing – review and editing. Janine Bosak:
Writing – original draft; writing – review and editing. Jennifer Bosson: Writing – original draft; writ-
ing – review and editing. Marija Branković: Writing – original draft; writing – review and editing.
Julius Burkauskas: Writing – original draft; writing – review and editing. Vladimíra Čavojová:
Writing – original draft; writing – review and editing. Sapna Cheryan: Writing – original draft; writing
– review and editing. Eunsoo Choi: Writing – original draft; writing – review and editing. Incheol
Choi: Writing – original draft; writing – review and editing. Carlos C. Contreras- Ibáñez: Writing –
original draft; writing – review and editing. Andrew Coogan: Writing – original draft; writing – review
and editing. Ivan Danyliuk: Writing – original draft; writing – review and editing. Ilan Dar- Nimrod:
Writing – original draft; writing – review and editing. Nilanjana Dasgupta: Writing – original draft;
writing – review and editing. Soledad de Lemus: Writing – original draft; writing – review and editing;
funding acquisition. Thierry Devos: Writing – original draft; writing – review and editing. Marwan
Diab: Writing – original draft; writing – review and editing. Amanda B. Diekman: Writing – original
draft; writing – review and editing. Maria Efremova: Writing – original draft; writing – review and
editing; funding acquisition. Léïla Eisner: Writing – original draft; writing – review and editing; fund-
ing acquisition. Anja Eller: Writing – original draft; writing – review and editing. Rasa Erentaite:
Writing – original draft; writing – review and editing. Denisa Fedáková: Writing – original draft; writ-
ing – review and editing; funding acquisition. Renata Franc: Writing – original draft; writing – review
and editing. Leire Gartzia: Writing – original draft; writing – review and editing. Alin Gavreliuc:
Writing – original draft; writing – review and editing. Dana Gavreliuc: Writing – original draft; writing
– review and editing. Julija Gecaite- Stonciene: Writing – original draft; writing – review and editing.
Adriana L. Germano: Writing – original draft; writing – review and editing. Ilaria Giovannelli:
Writing – original draft; writing – review and editing. Renzo Gismondi Diaz: Writing – original draft;
writing – review and editing. Lyudmila Gitikhmayeva: Writing – original draft; writing – review and
editing. Abiy Menkir Gizaw: Writing – original draft; writing – review and editing. Biljana Gjoneska:
Writing – original draft; writing – review and editing. Omar Martínez González: Writing – original
draft; writing – review and editing. Roberto González: Writing – original draft; writing – review and
editing; funding acquisition. Isaac David Grijalva: Writing – original draft; writing – review and edit-
ing. Derya Güngör: Writing – original draft; writing – review and editing. Marie Gustafsson Sendén:
Writing – original draft; writing – review and editing. William Hall: Writing – original draft; writing
– review and editing; funding acquisition. Charles Harb: Writing – original draft; writing – review and
editing. Bushra Hassan: Writing – original draft; writing – review and editing. Tabea Hässler: Writ ing
– original draft; writing – review and editing; funding acquisition. Diala R. Hawi: Writing – review
and editing; writing – original draft. Levke Henningsen: Writing – original draft; writing – review and
editing. Annedore Hoppe: Writing – original draft; writing – review and editing. Keiko Ishii: Writing
– original draft; writing – review and editing. Ivana Jakšić: Writing – original draft; writing – review
and editing. Alba Jasini: Writing – review and editing; writing – original draft. Jurgita Jurkevičienė:
Writing – original draft; writing – review and editing. Kaltrina Kelmendi: Writing – original draft;
writing – review and editing. Teri A. Kirby: Writing – original draft; writing – review and editing;
funding acquisition. Yoko Kitakaji: Writing – original draft; writing – review and editing. Natasza
Kosakowska- Berezecka: Writing – original draft; writing – review and editing. Inna Kozytska:
Writing – original draft; writing – review and editing. Clara Kulich: Writing – original draft; writing
– review and editing. Eva Kundtová- Klocová: Writing – original draft; writing – review and editing;
funding acquisition. Filiz Kunuroglu: Writing – original draft; writing – review and editing. Christina
Lapytskaia Aidy: Writing – original draft; writing – review and editing; funding acquisition. Albert
Lee: Writing – original draft; writing – review and editing. Anna Lindqvist: Writing – original draft;

|
19
PAREN TAL LEAVE POLICY AN D SOCI AL NOR MS
writing – review and editing. Wilson López- López: Writing – original draft; writing – review and edit-
ing. Liany Luzvinda: Writing – original draft; writing – review and editing. Fridanna Maricchiolo:
Writing – original draft; writing – review and editing. Delphine Martinot: Writing – original draft;
writing – review and editing. Rita Anne McNamara: Writing – original draft; writing – review and
editing. Alyson Meister: Writing – original draft; writing – review and editing. Tizita Lemma Melka:
Writing – original draft; writing – review and editing. Narseta Mickuviene: Writing – original draft;
writing – review and editing. María Isabel Miranda- Orrego: Writing – original draft; writing – review
and editing. Thadeus Mkamwa: Writing – original draft; writing – review and editing. James
Morandini: Writing – original draft; writing – review and editing. Thomas Morton: Writing – original
draft; writing – review and editing. David Mrisho: Writing – original draft; writing – review and edit-
ing. Jana Nikitin: Writing – original draft; writing – review and editing. Sabine Otten: Writing – origi-
nal draft; writing – review and editing. Maria Giuseppina Pacilli: Writing – original draft; writing
– review and editing. Elizabeth Page- Gould: Writing – original draft; writing – review and editing;
funding acquisition. Ana Perandrés-Gómez: Writing – original draft; writing – review and editing.
Jon Pizarro: Writing – original draft; writing – review and editing. Nada Pop- Jordanova: Writing –
original draft; writing – review and editing. Joanna Pyrkosz- Pacyna: Writing – original draft; writing
– review and editing. Sameir Quta: Writing – original draft; writing – review and editing. TamilSelvan
Ramis: Writing – original draft; writing – review and editing. Nitya Rani: Writing – original draft;
writing – review and editing. Sandrine Redersdorff: Writing – original draft; writing – review and
editing. Isabelle Régner: Writing – original draft; writing – review and editing. Emma A. Renström:
Writing – original draft; writing – review and editing. Adrian Rivera- Rodriguez: Writing – original
draft; writing – review and editing. Rocha- Sánchez Tania Esmeralda: Writing – original draft; writ-
ing – review and editing. Tatiana Ryabichenko: Writing – original draft; writing – review and editing;
funding acquisition. Rim Saab: Writing – original draft; writing – review and editing. Kiriko Sakata:
Writing – original draft; writing – review and editing. Adil Samekin: Writing – original draft; writing
– review and editing. Tracy Sánchez- Pacheco: Writing – original draft; writing – review and editing.
Carolin Scheifele: Writing – original draft; writing – review and editing. Marion K. Schulmeyer:
Writing – original draft; writing – review and editing. Sabine Sczesny: Writing – original draft; writing
– review and editing. David Sirlopú: Writing – original draft; writing – review and editing. Vanessa
Smith- Castro: Writing – original draft; writing – review and editing. Kadri Soo: Writing – original
draft; writing – review and editing. Federica Spaccatini: Writing – original draft; writing – review and
editing. Jennifer R. Steele: Writing – original draft; writing – review and editing; funding acquisition.
Melanie C. Steffens: Writing – original draft; writing – review and editing. Ines Sucic: Writing –
original draft; writing – review and editing. Joseph Vandello: Writing – original draft; writing – review
and editing. Laura Maria Velásquez- Díaz: Writing – original draft; writing – review and editing.
Melissa Vink: Writing – original draft; writing – review and editing. Eva Vives: Writing – original
draft; writing – review and editing. Turuwark Zalalam Warkineh: Writing – original draft; writing –
review and editing. Iris Žeželj: Writing – original draft; writing – review and editing. Xiaoxiao Zhang:
Writing – original draft; writing – review and editing; funding acquisition. Xian Zhao: Writing – origi-
nal draft; writing – review and editing. Yasin Koc: Writing – original draft; writing – review and edit-
ing. Ömer Erdem Kocak: Writing – original draft; writing – review and editing. Sarah E. Martiny:
Writing – original draft; writing – review and editing; conceptualization; methodology.
AFFILIATIONS
1Federal University of Applied Ad min strat ive Sciences, Berli n, Germany
2Leupha na Universit y, Lüneburg, Germany
3Inland Norway University of Applied S ciences, Li llehammer, Norway
4FernUniversität in Hag en, Germany
5University of Amsterda m, The Nether lands
6University of Leuven, Belgium
7The Universit y of British Columbia, Ca nada
8Thomas More College of Appl ied Psychology, Antwer p, Belg ium
9University of Arizon a, USA

20
|
SCHINDLER et al.
10University of Toronto, Canada
11Univer sity of Tar tu, Estonia
12Utrecht Un iversity, The Net herlands
13Slovak Ac ademy of Sciences, Bratislava, Slovaki a
14Australian Cat holic Un iversity, Melbou rne, Au stralia
15La Trobe Un iversit y, Austra lia
16State Islam ic Unive rsity Syarif Hidayat uillah Ja kar ta, I ndonesia
17Jag iellonian University, Krakow, Poland
18Bielefe ld University of Applied S cience s, Ger many
19Indiana Un iversity, Bloom ing ton, USA
20Rochester Institute of Technology in Kosovo (R IT), Kosovo
21University of Gene va, Swit zerland
22Univers ity of Waterloo, Canada
23Univers ity of Ka nsas, USA
24West Univers ity of Timisoa ra, Rom ania
25University of Ti rana , Alba nia
26Dubli n City Un iversity, Irel and
27Univer sity of South Flor ida, USA
28Singidunum University, Serbia
29Lithuanian Unive rsity of Health Sciences, Palanga , Lithuani a
30University of Washington, USA
31Korea University, South Korea
32Seou l Nation al Universit y, South Korea
33UAM – Universida d Autónoma Metropol itana, Mexico
34Maynoot h Univer sity ( National Uni versit y of Ireland), Ireland
35Taras Shevc henko Nat ional Univers ity of Kyiv, Uk raine
36The Universit y of Sydney, Aust ral ia
37University of M assachu setts , Amhe rst, Massachusetts, USA
38Universidad de Granada, Spa in
39San Die go State Univer sity, USA
40Stellenbosch Un iversity, Sout h Africa, Stellenbosch, Palesti ne
41Nation al Rese arch Un iversity Higher School of Econom ics (H SE University), Moscow, Russia
42Univer sity of Zu rich, Switzerland
43UNA M – Univer sidad Naciona l Autónoma de México, Me xico
44Kauna s Univer sity of Technology, Lit huania
45Inst itute of So cial S cience s Ivo Pilar, Croatia
46Deusto Business School, Bi lbao, Spain
47Yale Universit y, USA
48Univers ity ‘G. D'Annunzio’ of Chieti- Pescara, Italy
49Universidad Pr ivada de Santa C ruz de l a Sierr a, Bol ivia
50L. N. Gu mil yov Eurasian Nat ional Univers ity, Kazakhstan
51Bah ir Dar Un iversity, Eth iopia
52Macedonia n Academy of S ciences and Ar ts, Skopje, Nort h Macedonia
53Pontificia Universidad Católica de Chile, Chile
54Pontificia Universidad Católica del Ecuador, Ecuador
55Stock holm Universit y, Sweden
56Brock Universit y, Canada
57American University of Beir ut, Leb anon
58Internation al Islamic Un iversity Islamabad, Pakistan
59Doha Institute for Graduate Studies, Qatar
60University of Exeter, UK
61University of Le ipzig , Germ any
62Nagoya Un iversity, Japan
63University of Belgra de, Serbia
64University of Prishtina, Kosovo
65Purdue University, West Laf ayette , USA
66Hiroshima University, Japan
67University of Gdańsk, Poland
68Masar yk University, Br no, Czech Republic
69İzmi r Kâtip Çelebi University, Turkey
70York University, Cana da
71Nanyang Technolog ical Un iversity, Singapore
72Lund Un iversity, Sweden
73Pontif icia Universidad Javer iana , Colombia
74Roma Tre Un iversit y, Italy
75Université Cle rmont Auve rgne ( LAPSCO), France
76Victoria Un iversit y of Wellington, New Ze ala nd
77IMD Business School , Switzerland

|
21
PAREN TAL LEAVE POLICY AN D SOCI AL NOR MS
78Saint Augustine University of Tanzania, Tanzania
79University of Copenhagen, Denmark
80Univers ity of Vie nna, Austria
81University of Gron ingen, The Netherlands
82Università di Perug ia, Ita ly
83AGH University of Kr akow, Poland
84Islamic University of G aza, Pa lesti ne
85Sunway Un iversity, Malaysia
86Independent Researcher, India
87Aix Ma rseille Uni v, CNRS , LPC , France
88University of Gothenburg, Sweden
89University of Sussex, UK
90M. Nari kbayev KAZGU U University, Astana , Kazakhst an
91Osnabrück Univer sity, Germany
92Research Found ation of F lande rs, Bel gium
93RPTU Kaiserslautern- Landau, Landau, Germany
94University of Bern, Switzerl and
95Univer sidad San Sebastiá n, Conce pción, C hile
96University of Costa Rica, Cost a Rica
97Shenzhen University, China
98Ohio Universit y, USA
99Instanbul Med ipol Uni versit y, Turkey
100UiT T he Arctic Universit y of Norway, Nor way
ACKNOWLEDGEMENTS
This work was supported by a Grant of the German Research Foundation (DFG; Grant ID SCHU 3362/2-
1) to the second author. Furthermore, this research project was conceived following the award of a SSHRC
Insight Development Grant to Toni Schmader (430- 2018- 00361). Additional funding included: a SSHRC
Insight Grant awarded to J. R. Steele (435- 2014- 1247) and a SSHRC doctoral fellowship awarded to C.
Lapytskaia Aidy; funding from the Basic Research Program at HSE University, RF, awarded to Tatiana
Ryabichenko and Maria Efremova; a grant from the Economic and Social Research Council awarded to
Teri A. Kirby (ES/S00274X/1); funding from Spanish State Research Agency awarded to Soledad de Lemus
(PID2019- 111549 G B - I0 0/10.13039/501100011033); funding from Guangdong 13th- five Philosophy and
Social Science Planning Project (GD20CXL06) + National Natural Science Foundation of China awarded
to XiaoXiao Zhang (31600912); funding from the research infrastructure HUME Lab Experimental
Humanities Laboratory, Faculty of Arts, Masaryk University awarded to Eva Kundtová- Klocová; two grants
from the Swiss National Science Foundation awarded to Tabea Hässler (P1ZHP1_184553) and Léïla Eisner
(P500PS_206546); funding from the Center for Social Conflict and Cohesion Studies (ANID/FONDAP
#15130009) and the Center for Intercultural and Indigenous Research (ANID/FONDAP #15110006)
awarded to Roberto González; a SSHRC Postdoctoral Fellowship (756- 2017- 0249) awarded to William Hall;
funding awarded to Denisa Fedakova from the Slovak Research and Development Agency project (APVV
20- 0319); funding awarded to Léïla Eisner from the Swiss National Science Foundation (P2LAP1_194987)
and funding from Canada Research Chairs (CRC 152583), the Social Sciences and Humanities Research
Council (Insight Grant 140649), and the Ontario Ministry of Research and Innovation (Early Research
Award 152655) awarded to Elizabeth Page- Gould.
CONFLICT OF INTEREST STATEMENT
The authors declare no conflicts of interest.
DATA AVAILABILITY STATEMENT
This research was conducted consistent with open science practices. Exclusion criteria, hypotheses, and
analyses were registered prior to data analysis. Pre- registration, materials, data, and procedure are pub-
licly available on the Open Science Framework: https:// osf. io/ dzq3m/ ? view_ only= fa56c e28b7 92481
18981 bee6c e4c1db6.

22
|
SCHINDLER et al.
ORCID
Simon Schindler https://orcid.org/0000-0003-1764-7241
Janine Bosak https://orcid.org/0000-0001-5701-6538
Alin Gavreliuc https://orcid.org/0000-0001-8411-0327
Abiy Menkir Gizaw https://orcid.org/0000-0002-6528-0977
Biljana Gjoneska https://orcid.org/0000-0003-1200-6672
Annedore Hoppe https://orcid.org/0000-0003-4708-3301
Jana Nikitin https://orcid.org/0000-0003-1642-154X
Melanie C. Steffens https://orcid.org/0000-0002-7915-3629
Turuwark Zalalam Warkineh https://orcid.org/0000-0001-8121-0007
REFERENCES
Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179–211. https:// doi. org/
10. 1016/ 0749 - 5978(91) 90 02 0 - T
Armitage, C. J., & Conner, M. (2001). Eff icacy of the theor y of planned behav iour: A meta- analytic rev iew. British Journal of Social
Psyc holog y , 40, 471–499. https:// doi. org/ 10. 1348/ 01446 66011 64939
Bicchieri, C., & Mercier, H. (2014). Norms and bel iefs: How change occurs. In M. Xen itidou & B. Emonds ( Eds.), The com plexit y
of social norms (pp. 37–54). Springer.
Bishin, B. G., Hayes, T. J., Incantalupo, M. B., & Smith, C. A. (2016). Opinion backlash and publ ic att itudes: Are pol itical
advances in gay rights counterproductive? Amer ican Jour nal of Political Science, 60(3), 625 –648. h t tps :// doi. org / 10. 1111/ ajps .
12181
Boeckmann, I., Misra, J., & Budig, M. J. (2014). Cult ural and institutiona l factors shaping mothers' employment and working
hours in postindustr ial cou ntries. Social Forces, 93(4), 1301–1333. ht tps:// doi. org/ 10. 1093/ sf/ sou119
Bosson, J. K ., Wilkerson, M., Kosakowska-Berezecka, N., Jurek, P., & Olech, M. (2022). Harder won and easier lost? Testing the
double standard in gender rules in 62 countries. Sex Ro les, 87, 1–19. https:// doi. org/ 10. 10 07/ s11199- 022- 01297- y
Burgess, D. J., & Borgida, E. (1999). Who women are, who women should be: Descriptive and prescriptive gender stereotyping
in sex discrimination descriptive and prescript ive gender stereot yping. Psycholog y, Public Policy, and L aw, 5(3), 665– 692 .
htt ps:// doi. or g/ 10. 1037/ 1076- 8971. 5. 3. 665
Carlson, D. L., Hanson, S., & Fitzroy, A. (2016). The d ivision of child care, sexual inti macy, and relationship quality in couples.
Gender & Society, 30(3), 442–466. https:// doi. org/ 10. 1177/ 08912 43215 62670 9
Chung, A., & Rimal, R. N. (2016). Social norms: A review. Review of Communication Research, 4, 1–28. htt ps:// doi. org/ 10. 12 84 0/
issn. 2255- 4165. 2016. 04. 01. 008
Cialdin i, R. B., Kallgren, C. A., & Reno, R . R. (1991). A focus theory of normative conduct: A t heoretical refinement and reeval-
uation of the role of norms i n human behavior. In M. P. Zanna (Ed.), Advances in experimental social psycholog y (Vol . 24, pp.
201–234). Academic Press. https:// doi. org/ 10. 1016/ S0065 - 2601(08) 60330 - 5
Cialdin i, R. B., & Trost, M. R. (1998). Social influence: Social norms, conformity, and compliance. In D. T. Gilbert, S. T. Fiske,
& G. Lindzey (Eds.), The handbook of social psycholog y (pp. 151–192). McGraw- H il l.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Erlbaum.
Croft, A., Schmader, T., & Block , K. (2015). An underexamined inequalit y: Cultural and psychological barriers to men's en-
gagement with communal roles. Personalit y and Social Psycholog y Review, 19(4), 343–370. htt ps:// doi. org/ 10. 1177/ 1088 8 68314
564789
Croft, A., Schmader, T., & Block , K. (2019). Life in t he balance: Are women's possible selves const rained by men's domestic
involvement? Personality and Soc ial Psycholog y Bulletin, 45(5), 808 –823. htt ps:// doi. org/ 10. 1177/ 01461 67218 797294
de Visser, R . O., & McDonnell, E. J. (2013). “Man points”: Mascul ine capital and young men’s health. Health Psycholog y, 32(1),
5–14. ht tps:// doi. org / 10. 1037/ a0029045
Dotti Sani , G. M. (2020). Is it “good” to have a stay- at- home mom? Parental childcare t ime and work–family a rrangements in
Italy, 1988 –2014. Social Politics: International Studies in Gender, State and Society, 28(4), 896–920.
Dotti Sani , G. M., & Treas, J. (2016). Educationa l gradients i n parents' child- care time across countries, 1965–2012. Journal of
Marr iage and Family, 78(4), 1083–1096 .
Eagly, A. H., & Wood, W. (2012). Social role t heory. In P. A. M. Van Lange, A. W. Kruglanski, & E . T. Higg ins ( Eds.), Handbook
of theories of social psycholog y (pp. 458 –476). Sage. htt ps:// doi. org/ 10. 4135/ 97814 46249 222. n49
Eisner, L ., Turner- Zwinkels, F., & Spini, D. (2021). The impact of laws on norms perceptions. Personality and Social Psycholog y
Bulletin, 47(7), 1071–1083. ht t ps:// doi. org / 10. 117 7/ 01461 67220 959176
Enders, C. K., & Tofighi, D. (2 007 ). Center ing predictor variables in cross- sect ional mu lti level models: A new look at an old
issue. Psychological Methods, 12(2), 121–138. https:// doi. org/ 10. 1037/ 1082 - 989X. 12. 2. 121
Eriksson, K., Strimling, P., & Coultas, J. C. (2015). Bidirectional associations between descriptive a nd injunctive norms.
Organizational Behavior and Human Decision Processes, 129, 59– 69. https:// doi. org/ 10. 1016/j. obhdp. 2014. 09. 011
Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention, and behavior: An introduction to theory and research. Addison- Wesley.

|
23
PAREN TAL LEAVE POLICY AN D SOCI AL NOR MS
Flores, A. R., & Barclay, S. (2016). Backlash, consensus, legitimacy, or polarization: The effect of same- sex marriage policy on
mass att itudes. Political Research Quarterly, 69(1), 43–56. https:// doi. org/ 10. 1177/ 10659 12915 621175
Galbiati, R ., Henry, E., Jacquemet, N., & L obeck, M. (2021). How laws affect t he perception of norms: Empirical ev idence from
the lockdown. PLoS One, 16(9), e0256624. https:// doi. org/ 10. 1371/ journ al. pone. 0256624
Gaunt, R. (20 06). Biologica l essent ial ism, gender ideologies, and role att itudes: What determines parents’ involvement in child
care. Sex Roles, 55(7–8), 523–533. ht tps:// doi. org / 10. 1007/ s11199- 00 6- 9105 - 0
Goldstei n, N. J., & Ciald ini, R. B. (2007). The spyglass self: A model of vicarious self- percept ion. Journal of Personal ity and Social
Psyc holog y , 92, 402– 417. https:// doi. org/ 10. 1037/ 002 2- 3514. 92 .3. 402
Haas, L ., & Hwang, C. P. (2019). Workplace support a nd European fat hers' use of state policies promoting shared ch ildcare.
Communit y, Work & Family, 22(1), 1–22. ht tps:// doi. org / 10. 108 0/ 13668 8 03. 2018. 155620 4
Hami lton, W. L., Biener, L ., & Brennan, R . T. (2008). Do local tobacco reg ulations influence perceived smok ing norms?
Evidence from adult a nd youth su rveys in Massachusetts. Health Education Research, 23, 709–722. htt ps:// doi. org/ 10. 10 93/
he r / c y m 054
Hogg, M. A. (2010). Influence and leadersh ip. In S. T. Fiske, D. T. Gilbert, & G. Lindzey ( Eds.), The handbook of social psycholog y
( Vol . 2, 5th ed., pp. 1166–120 6). Wiley.
Hogg, M. A., & Reid, S. A. (2 006 ). Social ident ity, self- categorization, and t he communicat ion of group norms. Communication
Theory, 16(1), 7–30. https:// doi. org/ 10. 1111/j. 1468- 2885. 2006. 00003. x
Horne, C., & Mollborn, S. (2 020). Norms: An integrated framework. Annual Review of Sociolog y, 46, 467–487. htt ps:// doi. org/ 10.
1146/ annur ev- soc- 12191 9- 054658
Jurado- Guerrero, T., & Muñoz- Comet, J. (2021). Design matters most: Changing social gaps in the use of fathers' leave in Spain.
Population Research and Policy R eview, 40, 589–615. htt p s:// doi . or g/ 10 . 100 7/ s1111 3 - 0 2 0 - 09592 - w
Ki nzig, A. P., Ehrl ich, P. R., Alston , L. J., Arrow, K. , Barrett, S ., Buchman, T. G., Da ily, G. C., Lev in, B., Lev in, S., Oppe nheimer,
M., Ostrom, E., & Saari, D. (2013). Social norms and global env ironmental chal lenges: The complex interaction of behav-
iors, values, and policy. Bioscience, 63(3 ), 16 4 –175.
Korotayev, A., Bi lyuga, S., & Shishkina, A. (2018). GDP per capita and protest activity: A quant itative reanalysis. Cross- Cultural
Research, 52(4), 406–4 40. ht tps:// doi. org/ 10. 1177/ 10693 97117 73232 8
Laurin, K. (2018). Inaugurating rationalization: Th ree field stud ies find increased rationa lization when anticipated realities
become current. Psychological Science, 29(4), 483– 495. ht t p s:// doi. org / 10. 1177/ 0 95 67 97617 73 8814
Legros, S., & Cislaghi, B. (2020). Mapping the social- norms literat ure: A n overview of rev iews. Perspectives on Psychological Science,
15(1), 62– 80. h ttps :// doi . or g/ 10. 1177/ 17456 91619 86 6 45 5
Luís, S., & Palma- Oliveira, J. (2016). Public policy and social norms: The case of a nationwide smoking ba n among college
students. Psycholog y, Public Policy, and Law, 22(1), 22–30. https:// doi. org/ 10. 1037/ law00 00064
Mahmoud, M. A., Ahmad, M. S., Yusoff, M. Z. M., & Mustapha, A. (2 014). A review of norms and normative multiagent sys-
tems. The Scientif ic World Journal, 2 014, 684587. https:// doi. org/ 10. 1155/ 2014/ 684587
McAdams, R. H. (200 0). An attitudinal theory of expressive law. Oregon Law Review, 79, 339–390.
Meeussen, L., Van Laar, C., & Van Grootel, S. (2020). How to foster male engagement in traditional ly female communal roles
and occupations: Insights from research on gender norms and preca rious manhood. Social Issues and Policy Review, 14 (1),
297–328. ht t ps:// doi. o rg/ 1 0. 1111/ si pr. 12 0 60
Meeussen, L., Van Laar, C., & Verbruggen, M. (2 019). Looking for a family man? Nor ms for men are toppli ng in heterosexual
relationships. Sex R oles, 80(7), 429–442. ht tps:// doi. org / 10. 1007/ s11199- 018- 0946 - 0
Meeussen, L., Veldman, J., & Van Laar, C. (2016). Combining gender, work, and family identities: T he cross- over and spill-over
of gender norms into young adults’ work and family aspirations. Frontiers in Psychology, 7, 1–11. http s:// doi. org / 10. 3389/
fpsyg. 2016. 01781
Morris, M. W., Hong, Y.- y., Chiu, C.- y., & Liu, Z. (2015). Normology: Integrating insights about social norms to understand
cultural dynam ics. Organizational Behavior and Human Decision Processes, 129, 1–13. https:// doi. org/ 10. 1016/j. obhdp. 2015.
03. 001
Nyborg, K . (2003). The impact of public policy on socia l and moral norms: Some examples. Journal of Consumer Policy, 26(3),
259–2 77. https:// doi. org/ 10. 1023/A: 10256 22223207
Olsson, M., van Grootel, S., Block, K ., Schuster, C., Meussen, L ., Schmader, T., Croft, A., Su n, M. S., Ainsaar, M., Aarntzen,
L., Adamus, M., Anderson, J., Atkinson, C., Avicenna, M., Bąbel, P., Barth, M., Benson- Greenwald, T. M., Maloku, E.,
Berent, J., … Martiny, S. E. (2023). Gender gap in parenta l leave intentions: Evidence from 37 countr ies. Political Psycholog y,
44, 1163 –1192 . ht t ps :// doi. org / 10. 1111/ po ps. 12 8 80
Omidakhsh, N., Sprague, A., & Heyma nn, J. (2020). Dismant ling restrictive gender norms: Can better designed paternal leave
policies help? Analyses of Social Issues and Public Policy, 20(1), 382– 396 .
Orbell, S., Lidierth, P., Henderson, C. J., Geeraert, N., Uller, C., Uskul, A. K., & Kyriakaki, M. (2009). Social–cognitive beliefs,
alcohol, and tobacco use: A prospective community study of change follow ing a ban on smoking in public places. Health
Psyc holog y , 28(6), 753 –761. htt ps:// doi. org/ 10. 1037/ a0016943
Pailhé, A., Solaz, A ., & Stanfors, M. (2021). The great convergence: Gender and u npaid work in Europe and the United States.
Population and Development Review, 47(1), 181–217.
Posner, E. (2000). Law and social norms. Harvard University Press.

24
|
SCHINDLER et al.
Prentice, D. A., & Carranza, E. (2002). What women and men should be, shouldn't be, are allowed to be, and don't have to be:
The contents of prescriptive gender stereotypes. Psycholog y of Women Quarterly, 26(4), 269–281. h ttps :// doi . or g / 10 . 1111/
1471- 6402. t01- 1- 000 66
Rivis, A., & Sheeran, P. (2003). Descriptive norms as an addit ional predictor in the theory of planned behaviour: A meta-
analysis. Current Psycholog y, 22(3), 218–233. https:// doi. org/ 10. 100 7/ s1214 4 - 003- 1018- 2
Rudman, L. A ., & Fairchi ld, K. (2004). React ions to cou nterstereoty pic behavior: The role of back lash in cultural stereotype
maintenance. Journal of Personality and Social Psycholog y, 87, 157–176. https:// doi. org/ 10. 1037/ 0022- 3514. 87.2. 157
Schwar tz, S. H. (20 08). The 7 Schwar tz cu ltural value orientation scores for 80 countries. https:// doi. org/ 10. 13140/ RG.2. 1.
3313. 3040
Schwar tz, S. H. (2014). Rethin king the concept and measurement of societal culture i n light of empirical findings. Journal of
Cross- Cultural Psycholog y, 45(1), 5–13. https:// doi. org/ 10. 1177/ 00220 22113 490830
Steinbach, A., & Schulz, F. (2022). Stability a nd change in German parents' childcare patterns across t wo decades. Social Polit ics:
International Studies in Gender, State and Society, 29(2), 428– 445.
Sunstein, C. R. (1996). On the expressive function of law. University of Pennsylvania Law Review, 144 (5), 2021–2053.
Syropoulos, S., Sparkman, G., & Constantino, S. M. (2024). The ex pressive f unct ion of public policy: Renewable energ y mandates
signal social norms. Philosophical Transactions of the Royal Society B, 379, 20230038. https:// doi. org/ 10. 1098/ rstb. 20 23. 0038
Tankard, M. E., & Paluck, E. L. (2016). Norm perception as a vehicle for social change. Social Issues and Policy Review, 10(1),
181–211. htt ps:// do i. o rg/ 1 0. 1111/ si pr. 12 022
Tankard, M. E., & Paluck, E. L. (2017). The effect of a supreme court decision regarding gay marriage on social norms and
persona l attitudes. Psychological Science, 28(9), 1334–13 44. https:// doi. org/ 10. 1177/ 09567 97617 709594
Tharp, D. T., & Parks- Stamm, E. J. (2021). Gender differences in the intended use of parental leave: Implications for human
capital development. Journal of Family and Economic Issues, 42, 47– 60. ht tps:// doi. org/ 10. 10 07/ s1083 4 - 020 - 09722 - 8
Unterhofer, U., & Wrohlich, K. (2017). Fat hers, parenta l leave and gender nor ms (I ZA Discussion Papers). https:// papers. ssrn.
com/ Sol3/ papers. cfm? abstr act_ id= 2952289
Van Kleef, G. A ., Gelfand, M. J., & Jetten, J. (2019). The dynam ic natu re of social norms: New perspectives on norm devel-
opment, impact, violat ion, and enforcement. Journal of Experimental Social Psychology, 84, 103814. http s:// doi. org / 10. 1016/j.
jesp. 2019. 05. 002
Wei, L. (2020). Trends i n parental t ime allocated to chi ld care: Evidence from Canada, 1986 –2010. Canadian Public Policy, 46(2),
236 –252 .
SUPPORTING INFORMATION
Additional supporting information can be found online in the Supporting Information section at the
end of this article.
How to cite this article: Schindler, S., Schuster, C., Olsson, M. I. T., Froehlich, L., Hübner,
A.-K., Block, K., Van Laar, C., Schmader, T., Meeussen, L., van Grootel, S., Croft, A., Sun, M.
S., Ainsaar, M., Aarntzen, L., Adamus, M., Anderson, J., Atkinson, C., Avicenna, M., Bąbel, P.,
… Martiny, S. E. (2024). Policy as normative influence? On the relationship between parental
leave policy and social norms in gender division of childcare across 48 countries. British Journal of
Social Psycholog y, 00, 1–24. htt ps://doi.org/10.1111/bjs o.128 0 6