Revealing Gender Double Standards in the Parenthood Norm Depends on Question Order

Becoming a parent has been described as a dominant social norm, especially for women. Though some research has indicated changes toward more flexible gendered parenthood norms, methodological issues may be masking the continued presence of a gender double standard. In line with the condition for activation of double standards, we postulated that endorsement of the parenthood norm would vary depending on the response context. Our aim was to analyze the parenthood norm for women and for men taking into account the response context in a quantitative survey. In a French nationally-representative sample, more than 4,000 female and male adults were asked whether a woman/man can have a fulfilled life without having children in two questions presented in a random order. Based on the literature on question-order effects, the answer to the first question should be influenced by the participant’s personal background (e.g., gender, parental status), i.e., the personal background context, whereas the question asked second should be influenced by the comparison with the first question, i.e. the social relational context. In the personal background context, the own-gender parenthood norm was endorsed more strongly than the other-gender parenthood norm by both female and male participants. In contrast, in the social relational context, the parenthood norm for women was endorsed more strongly than the parenthood norm for men by both female and male participants. Our results showed a strong gender double standard observed only in the comparative context and illustrates the need to use appropriate survey methodology to examine the presence of gendered social norms.

Across the world, common lay beliefs emphasize that parenthood is essential to have a meaningful and fulfilled life (Hansen, 2012), and becoming a parent has been described as part of a "normal, expectable, life" (Neugarten, 1969, p. 125). Previous gender studies showed that this social norm of parenthood is highly gendered, with higher expectations for women than for men (Peterson, 2017). Though some research has indicated changes toward more flexible gendered parenthood norms (Hansen, 2012;Preisner et al., 2020;Rijken & Merz, 2014), the context in which the questions about the parenthood norm are answered may be masking the continued presence of a gender double standard. The aim of the current study is to examine the endorsement of the parenthood norm for women and for men among female and male participants in varied response contexts to determine whether a gender double standard exists that would otherwise go undetected.
As fatherhood is not perceived as a basic biological instinct, childless men are not exposed to the same social stigmatization as childless women (Gotman, 2017;Peterson, 2017;Tanaka & Johnson, 2016). Peterson (2015) drew a parallel between the childfree women she interviewed in Sweden and the childfree men interviewed by Terry and Braun in New Zealand (Terry & Braun, 2012). This study revealed that women's childlessness is more likely than men's to be attributed to selfishness and a desire for freedom (Peterson, 2015). Women also experience more social pressure to have children (Koropeckyj-Cox & Pendell, 2007b;Rozée & Mazuy, 2012;Thornton & Young-DeMarco, 2001).
At odds with this large literature on the gendered parenthood norm, Rijken and Merz (2014) examined data from the European Social Survey and found that attitudes toward men who chose not to have children were more negative than attitudes toward women who chose to remain childless in countries with high gender equality, including France, the United Kingdom, Germany, Sweden, and Spain, suggesting a "double standard favouring women" (Rijken & Merz, 2014, p. 479). The authors proposed that the more tolerant attitude toward childless women might reflect the higher social cost of parenthood for women than for men, with "strains and sacrifices in personal and professional life" attached to motherhood (Rijken & Merz, 2014, p. 472). However, in this study, each participant was only interviewed on one item: half of the participants were interviewed on the norm for women (but not for men) and half on the norm for men (but not for women). Drawing from research on the double standards framework, we consider whether this response context may have masked an otherwise observable gendered parenthood norm.

Double Standards Framework
Differences based on status characteristics (e.g., gender) have been conceptualized in the theory of status difference as "double standards" (Foschi, 2000). The "double standards" concept denotes the fact that the same behavior (e.g., childlessness) is considered and judged differently depending on the social category of the individual (e.g., gender, ethnicity, socioeconomic class). This cognitive bias leads to stricter standards for the person whose social category holds a lower status (Risman, 2004). More lenient standards for higher status persons are a "reward" of their higher status, and also function to maintain status differences (Foschi, 2000).
The double standards concept makes it possible to understand and analyze gender inequalities in different areas. In the professional sphere, the double standards concept explains why women have to work harder and are allowed fewer mistakes than men at the same level of ability as their skills and actions are judged with stricter criteria than men (Foschi, 2000). Double standards have also been explored in the context of sexual behavior to explain why promiscuity is considered more socially acceptable for men than for women (Crawford & Popp, 2003;Milhausen & Herold, 1999). Gender norms dictate that women must appear "pure and virginal" to avoid being labelled as "promiscuous and easy" (Crawford & Popp, 2003). To conform to these social norms, women may, for example, refuse to carry condoms and instead rely on men to provide protection in sexual encounters, despite the health and contraceptive risks (Crawford & Popp, 2003).
A comprehensive review identified a set of methodological requirements to detect double standards (Crawford & Popp, 2003). First, the norm (or phenomenon) under investigation must be assessed using gendered questions (Crawford & Popp, 2003). This requirement was not met in most quantitative surveys carried out on the parenthood norm more than 25 years ago in the U.S., which included gender-neutral questions such as "people who have never had children lead empty lives" (American General Social Surveys of 1988 and1994;Koropeckyj-Cox & Pendell, 2007b;Thornton & Young-DeMarco, 2001), and "it is better to have a child than to remain childless" (American National Survey of Families andHouseholds of 1987-1988;Koropeckyj-Cox & Pendell, 2007a, 2007b). In the study described above using the European Social Survey (Rijken & Merz, 2014), gendered questions were used to examine the parenthood norm. Specifically, half of the participants responded to the following questions: "how much do you approve or disapprove if a woman chooses never to have children?" and half responded to "how much do you approve or disapprove if a man chooses never to have children?" (Rijken & Merz, 2014). However, gendered questions are not sufficient to detect gender double standards.
A within-subject design is also required to activate double standards (Crawford & Popp, 2003, p. 15). In a withinsubject design, participants are interviewed about the norm for women and the norm for men. As noted above, in the European survey, each participant was only interviewed on one item: half of the participants were interviewed on the norm for women (but not for men) and half on the norm for men (but not for women) (Rijken & Merz, 2014). This crucial point is detailed in the next section to clarify why double standards can only be observed using a within-subject design.

Activating Double Standards Through the Question-Order Effect
The use of a within-subject design is crucial because double standards may not emerge spontaneously. Foschi (2000) demonstrated that double standards can be revealed only when the study complies with the condition of activation of double standards. This condition of activation requires a comparative context. For example, Foschi observed no gender differences in the requirements for professional qualifications in the absence of a direct comparison being made between men and women, whereas stricter requirements for women emerged when men and women were directly compared. The author concluded that the process of making a comparison between women and men activates a double standard.
In line with the approach above, Ridgeway and Correll (2004) argued for an interactional approach to explore double standards. They defined a "social relational context" as a response context in which "individuals define themselves in relation to others" (Ridgeway & Correll, 2004, p. 511). This self-definition typically relies on a gender categorization that is socially constructed as binary: being a woman or being a man. When compared with the other gender, individuals moderate or exaggerate their responses according to their level of endorsement of gender norms. To investigate double standards, it is therefore necessary to develop a withinsubject design in which the participant is asked the same question about men and about women in order to be placed in a "social relational context" on the second question.
The underlying assumption is that the participant is not in the same context when answering the first and the second question and these two contexts lead to differences in the answer to the question. This effect is described in the survey methodology literature as the question-order effect (Moore, 2002). The question-order effect is not specific to gender-related questions; it has been observed more broadly in questions exploring personal opinions (Hayes, 1964;McFarland, 1981;Schuman & Presser, 1981). To answer a question about a personal opinion (such as the parenthood norm), participants pick a point of comparison or standard by which they evaluate the item (Moore, 2002). To illustrate, consider Moore's example asking whether Bill Clinton was trustworthy as a U.S. president. When directly questioned about Bill Clinton, the standard of comparison is shaped by elements such as participants' personal history, values, memories from news and discussions with relatives. Thus, participants were responding within a "personal background context." If the question about Bill Clinton was preceded by a question about whether Al Gore was trustworthy as a U.S. president, given the proximity of the two questions, it is likely that people would make their assessment of Bill Clinton in comparison with Al Gore. As the standard of comparison has evolved, this may have an impact on the response reflecting the social relational comparison with Al Gore. Thus, here, participants answered in a "social relational context." This question-order effect should not be considered a survey artefact but as a meaningful way to "make the norm more explicit" and to exhibit "the forces that shape responses in ordinary social interaction" (Schuman & Ludwig, 1983, p. 115 & 119).

Current Study
The objective of this paper is to explore gender double standards in the parenthood norm among French female and male adults taking into account the response context in a quantitative survey.
Consistent with past research on gender norms (Gotman, 2017;Peterson, 2017;Tanaka & Johnson, 2016), we anticipated a greater endorsement of the parenthood norm for women than for men. In line with the methodological framework on double standards (Crawford & Popp, 2003;Foschi, 2000), we assumed that the gendered parenthood norm would emerge only when conditions of activation of a double standard are met. For quantitative surveys, these conditions of activation have been described as the question-order effect (Moore, 2002), whereby the response to the first question affects the response to the second question. The response to the first question is shaped by personal history, values and environment, referred to as the "personal background context," whereas the response to the second question is shaped by the social relational comparison with the response to the first question, referred to as the "social relational context." We used these two different response contexts as a meaningful way to activate and thus reveal gender double standards in the parenthood norm.
Specifically, we expected that endorsement of the parenthood norm for women and for men among female and male participants would be different in the personal background context compared to the social relational context, revealing a question-order effect (Hypothesis 1). We also expected that gender double standards in the form of higher endorsement of the parenthood norm for women than for men would be observed among female and male respondents in the social relational context and not in the personal background context (Hypothesis 2). Finally, we examined whether the pattern of results would hold when controlling for participants' socioeconomic status, family profile, and life values.

Participants
Participants were contacted by phone using random-digit dialling in order to constitute a national representative sample of adults of reproductive age living in France in 2010. To take part in the Fecond survey (for detailed information on the survey, see Legleye et al., 2013), participants had to be aged between 15 and 49 years old. Participants were randomly divided into two ballots (J1 and J2), each one having specific questions to limit the length of the interview to 40 min. Half the participants (n = 4,261) answered to the J1 ballot that included questions on the parenthood norm. Five percent of these participants (n = 194 out of 4,261) were excluded because they failed to respond to at least one of the questions that were used in the present study. The results are based on a sample of 4,067 participants. Table 1 presents descriptive statistics for the sample. First, we selected three variables that reflected how close the participant was to parenthood: age at interview, being in a relationship, and parental status. Age at interview is considered an indicator of proximity to parenthood because in France the majority of people have children between the ages of 25 and 34 years (Breton et al., 2017). In our study population, nearly half (47%) of the participants were aged 35 years and older, 26% were aged 25-34 years and 27% were younger than 25 years. The majority stated that they had a partner (68%). Nearly half of the participants (49%) had no children at the time of the survey. Second, we used two variables to measure the social class of the participant: education and occupation. In our study population, 20% had a bachelor's degree, 50% had a lower level of education and 30% a higher degree. Lastly, we took into account the importance of religion in the participant's life (whatever the religion) as this may reflect some of their life values (Koropeckyj-Cox & Pendell, 2007a;Merz & Liefbroer, 2012;Rijken & Merz, 2014;van de Kaa, 1987). For half of the participants (49%), religion was not important in their life or they had no religion at all, whereas 6% of participants declared that religion was very important in their life.

Measures
In the Fecond survey, participants were interviewed about their sexual and reproductive life Moreau & Bohet, 2016;Moreau et al., 2014). The interview included 12 sections in the following order: sociodemographic profile, socialization, fertility intentions, reproductive biography, infertility care, contraceptive biography, focus on recent contraceptive use, first sexual intercourse, sexuality, norms, health, and health care.

Participant Gender
Participants indicated their gender in binary terms with the following question "are you … a male? A female?" Participants who defined themselves as female were classified as "female participants" and those who defined themselves as male were classified as "male participants."

Parenthood Norm
The parenthood norm was investigated using a gendered question in order to distinguish the parenthood norm for women and men. The parenthood norm for women was measured using the following question: "In your opinion, can a woman have a fulfilled life without having children?" The parenthood norm for men was measured with the same question adapted for men: "In your opinion, can a man have a fulfilled life without having children?" [translation of the French version: à votre avis, une femme/un homme peutelle/peut-il réussir sa vie sans avoir d'enfant ?]. The concept of a "fulfilled life" was not defined so that participants would be free to define it according to their own life values and purpose. Similarly, the concept of "having children" was not defined so that participants were free to define it in as a biological notion only (to conceive a child) or also as a a Statistics were weighted in order to take into account the two-stage stratified sampling design. The Fecond weighting integrates the probability of being selected in the sample (sampling weight). It also integrates a post-stratification adjustment to reflect the sociodemographic structure of the French population using French census data sociological notion (to parent a child, including stepchildren, adopted children, etc.). Participants selected one of the following response options: "Yes, easily," "With difficulty," or "No" [translation of the French version: Oui facilement, Difficilement, Non]. Participants were classified as endorsing the parenthood norm when they answered that a woman/man cannot, or can only with difficulty, have a fulfilled life without children. Participants were classified as not endorsing the parenthood norm when they answered that a woman/man can easily have a fulfilled life without having children.

Procedure
Consistent with recommendations for investigating double standards (Crawford & Popp, 2003), the Fecond Survey used a within-subject design: each participant was interviewed on the parenthood norm for women and men. Survey guidelines recommend handling question-order effects by randomly splitting the participants into two groups (Moore, 2002;Perreault, 1975): one group answers question A and then question B (first ballot) and the second group answers question B and then question A (second ballot). This is called a split-ballot design. This methodology was applied in the Fecond survey: half of our study population were first asked about the parenthood norm for women and then immediately asked about the norm for men (first ballot), whereas the other half were first asked about the parenthood norm for men and then immediately asked about the norm for women (second ballot). Participants are responding within a personal background context for the first question (i.e. based on personal history, values and environment) whereas they are responding within a social relational context for the second question (i.e. in comparison to the first question).

Analysis Strategy
All analyses were performed using Stata 13.0 software (StataCorp). Proportions were compared using chi-square tests. We expected that double standards would be activated only in the social relational context through the question-order effect. A first preliminary analysis was conducted to determine whether there was a questionorder effect (Hypothesis 1), i.e. if responses to the parenthood norm for women and for men among female and male respondents varied between the two response contexts, the personal background context and the social relational context. Then, we examined whether gender double standards emerged only in the social relational context and not in the personal background context among female and male respondents (Hypothesis 2). First, in the social relational context (i.e., responses to the second question), we examined whether gender double standards emerged in line with our hypothesis of a stronger endorsement of the parenthood norm for women than for men among female and male participants. Next, we examined the responses from the personal background context (i.e., responses to the first question) to determine if a different pattern emerged and if so, what was the logic of this pattern.
Finally, stratified multivariate analyses were conducted to determine whether the univariate analysis was affected by confounding variables. The dichotomous dependent variable was endorsement of the parenthood norm. Based on the literature (Merz & Liefbroer, 2012), we included six potential confounding variables that might affect endorsement of the parenthood norm: family profile (age, having a partner, having children), socioeconomic status (education, occupation) and life values (importance of religion for the participant). Results of the multivariate models give the odds ratio (OR), a measure of the association between a factor and an outcome (i.e., the parenthood norm). If the 95% confidence interval for the odds ratio does not include 1, then the association is statistically significant.

Parenthood Norm Across the Two Response Contexts
Parenthood norms for women and for men among female and male participants are presented in Table 2 across the two response contexts. We observed that the parenthood norm varied to a larger extent in the social relational context (from 37 to 66%) than in the personal background context (from 41 to 54%). Consistent with Hypothesis 1, the endorsement of the parenthood norm for women and for men among female and male participants varied with the response context (p-values ranging from .001 to .03). The variation in responses revealed that the endorsement of the parenthood norm for women was higher in the social relational context than in the personal background context among female and male participants, whereas endorsement of the parenthood norm for men was lower in the social relational context than in the personal background context among female and male participants.

Parenthood Norm in the Social Relational Context
In the social relational context, we hypothesized that a double standard would emerge such that the parenthood norm for women would be endorsed more strongly than the parenthood norm for men among female and male participants (Hypothesis 2). As shown in Table 2, female participants endorsed the parenthood norm for women to a greater degree than the parenthood norm for men (66% versus 40%, p = .001). Male participants also endorsed the parenthood norm for women more strongly than the norm for men (62% versus 37%, p = .001). For ease of reading, these proportions are presented in Fig. 1a. In the social relational context, female and male participants endorsed the parenthood norm for women (66% and 62%, p = .12) and men (37% and 40%, p = .23) to the same degree. In line with Hypothesis 2, we observed a gender double standard in the form of significantly stronger endorsement of the parenthood norm for women than for men by female and male participants in the social relational context.

Parenthood Norm in the Personal Background Context
In the personal background context, we observed a very different pattern of responding compared to the social relational context, consistent with Hypothesis 2. Among male participants, the parenthood norm for men was endorsed more strongly than the parenthood norm for women (52% versus 41%, p = .001). Thus, male participants endorsed more strongly the parenthood norm for their own gender than for the other-gender. A similar pattern was observed among female participants who also endorsed more strongly the parenthood norm for their own-gender (norm for women) than the parenthood norm for the other-gender (54% versus Personal background context refers to the first question in a non-comparative context. Social relational context refers to the second question in a comparative context. Gap refers to the difference between the two contexts in favor of the social relational context a Statistics were weighted in order to take into account the two-stage stratified sampling design. The Fecond weighting integrates the probability of being selected in the sample (sampling weight). It also integrates a post-stratification adjustment to reflect the sociodemographic structure of the French population using French census data b p value for chi-square test comparing the norm as a function of the question-order  Fig. 1b. There was no significant difference between female and male participants in the endorsed of the parenthood norm for their own gender (54% versus 52%, p = .30) or for the parenthood norm for the other gender (45% versus 41%, p = .15). This own-gender / other-gender pattern indicates that participant gender shaped endorsement of the parenthood norm in the personal background context. This pattern contrasts with the pattern of gender double standards observed in the social relational context.

Stratified Multivariate Analysis
Multivariate analyses across the two response contexts are presented in Table 3. After controlling for participant socioeconomic status, family profile, and life values, endorsement of the parenthood norm presented a very similar pattern in the multivariate analysis to the pattern described above. For ease of reading, odds ratios are presented in Fig. 1a for the social relation context and in Fig. 1b for the personal background context. In the social relational context, both female and male participants endorsed the parenthood norm for women more strongly than the parenthood norm for men. In the personal background context, female and male participants endorsed the parenthood norm for their own gender more strongly than the norm for the other gender.
In the univariate analyses comparing frequencies with chi-square tests (see left and right sides of Table 3), older participants, those who had a partner, and those who had children tended to endorse the parenthood norm more strongly. In the multivariate analysis, only the last two variables had a significant effect. The effect of age was no longer significant, perhaps because it was confounded with parental status (as older participants are also more often parents). The level of education of the participant was not significantly associated with endorsement of the parenthood norm. The occupation of the participant was significantly associated with norm endorsement in all analyses: in the multivariate analyses, participants who were executives had the lowest level of parenthood norm endorsement. Finally, the greater the importance of religion in the participants' life, the more often they endorsed the parenthood norm in both contexts.

Discussion
In the current study we explored endorsement of the parenthood norm for women and men among female and male participants, and examined differences based on the response context to test for gender double standards. In the social relational context, we observed a gender double standard favoring men with greater endorsement of the parenthood norm for women than for men among female and male participants. This finding is consistent with previous studies on gendered norms (Gotman, 2017;Peterson, 2017;Tanaka & Johnson, 2016), and contrasts with findings from the study described earlier based on the European Social Survey (Rijken & Merz, 2014). We observed that the parenthood norm for women was endorsed as strongly by female and male participants. Likewise, the parenthood norm for men was endorsed similarly by female and male participants. These patterns indicate that female and male participants shared the same social context in which parenthood is a gendered social construction. In the personal background context, we did not observe a gender double standard, illustrating the importance of conditions for the activation of double standards (Foschi, 2000). It clearly demonstrates how the same question in quantitative surveys can be answered differently depending on the context in which individuals are responding.
The gap in endorsement of the parenthood norm for women and for men was very large in the social relational context (26% among female participants and 25% among male participants) compared to the personal background context (9% among female participants and -11% among male participants). Based on Moore's framework, this larger gap in the social relational context reveals that female and male participants strongly contrast the parenthood norm for women and men. A contrast effect emerges when participants consider that the two items should be differentiated: they emphasize this difference in response to the second question and the gap between both items is larger than in response to the first question. Conversely, when participants consider that both items are close to each other, they emphasize this proximity in their response to the second question and the gap between both items is smaller than in response to the first question (this is known as the "consistency effect"). By observing a contrast effect, we demonstrate a gender double standard in the parenthood norm, as both female and male participants contrasted the norm for women and men.
In the personal background context, instead of endorsing more strongly the parenthood norm for women than for men, participants endorsed more strongly the parenthood norm for their own gender than for the other gender. It is noteworthy that in the personal background context, male participants endorsed the own-gender parenthood norm as strongly as female participants did. Our results showed that in this response context, the parenthood norm was primarily interpreted according to the participant's own gender. In other words, the personal elements used to define the standard to evaluate the parenthood norm principally reflect the participant's own gender. This own-gender / other-gender pattern is similar to the one exhibited in another field, studies of racism (Wilson, 2010) that explore perceptions of interracial prejudice. To reveal interracial prejudice among Black and White people, Wilson claimed that the analysis should not oppose "Black" and "White" questions but "in-group" ("White" for White people and "Black" for Black people) and "out-group" ("White" for Black people and "Black" for White people) questions. Similarly, we observed here an "in-group" (own-gender) and "out-group" (other-gender) response pattern. Personal background context refers to the first question in a non-comparative context. Social relational context refers to the second question in a comparative context a Univariate frequency was weighted in order to take into account the two-stage stratified sampling design. The Fecond weighting integrates the probability of being selected in the sample (sampling weight). It also integrates a post-stratification adjustment to reflect the sociodemographic structure of the French population using French census data b This variable crosses two dimensions: the parenthood norm (for women/for men) and the participant's gender (female/male). It has four modalities: female participants questioned on parenthood norm for women, female participants questioned on parenthood norm for men, male participants questioned on parenthood norm for women, and male participants questioned on parenthood norm for men

Other Factors Related to Endorsement of the Parenthood Norm
The effect of the control variables used in our multivariate analysis was identical in both contexts and did not modify the gender effects described in the univariate analysis. The associations observed between these variables and the parenthood norm were consistent with past studies. First, participants with a "family profile" (having children and/or a partner) endorsed the parenthood norm more strongly. It is important to note that this association does not allow causal conclusions, as this is only a cross-sectional survey. On the one hand, individuals who do not have a family profile may consider that they have a fulfilled life without having children.
On the other hand, non-endorsement of the parenthood norm could lead to childlessness being more frequent. These two directional interpretations of the association are not mutually exclusive, as both can coexist in a given society and even in the same individual. To interpret this association, a longitudinal approach and qualitative interviews are needed. Second, socioeconomic status (measured by education and occupation) was associated with endorsement of the parenthood norm such that higher social class participants tended to have a lower level of endorsement. Other studies have tended to show an even stronger effect of social class (Beaujouan et al., 2017;Koropeckyj-Cox & Pendell, 2007a;Merz & Liefbroer, 2012;Rijken & Merz, 2014). Such an effect of socioeconomic status is consistent with demographic statistics showing a higher level of childlessness among women with a university education compared with less educated women (Beaujouan et al., 2017). This socioeconomic gradient may reflect the fact that more highly educated individuals hold less traditional family attitudes than those with a low level of education (Merz & Liefbroer, 2012, p. 589). Following the same logic, women with a high level of education or professional position may be more likely to consider a fulfilling life independent of motherhood than women with less education or a less-valued profession.
Lastly, the life values indicator (importance of religion for the participant) was strongly associated with endorsement of the parenthood norm. The impact of religiosity has also been consistently observed in other studies (Koropeckyj-Cox & Pendell, 2007a;Merz & Liefbroer, 2012;Rijken & Merz, 2014). The majority of religions convey traditional values and often resist new behaviors of the Second Demographic Transition (Merz & Liefbroer, 2012;van de Kaa, 1987), including childlessness.

Limitations and Future Research Directions
The reliability and validity of the questions used in our study to assess the parenthood norm should be further considered. To date, very few large quantitative studies have explored the parenthood norm and they have all used different questions. In our study, we defined the parenthood norm as the ability to "have a fulfilled life without children." Two previous surveys explored the reverse side of this question: having an "empty" life without children (American General Social Surveys of 1998 and1994;Koropeckyj-Cox & Pendell, 2007b;Thornton & Young-DeMarco, 2001). Two surveys explored opinions on childlessness without focusing on life: "it is better to have a child than to remain childless" (American National Survey of Families andHouseholds of 1987-1988;Koropeckyj-Cox & Pendell, 2007a, 2007b and one explored approval/disapproval of the choice to never have children (European Social Survey of 2006; Rijken & Merz, 2014). These different approaches have never been validated. In addition, future research could explore whether women and men are conscious of their gendered approach to parenthood. It would be interesting to explore whether childlessness is perceived as a voluntary or involuntary phenomenon and what differentiates perceptions of voluntary and involuntary childlessness.
Another limitation is that this survey covers only one country. France has a strong policy of economic and infrastructure support to help its citizens to combine a family and a professional career. This may have an impact on gender norms such as the parenthood norm, if we consider gender as a system or a social structure (Ridgeway & Correll, 2004;Risman, 2004). In Germany, exploring the Family Panel (from 2008 to 2015), Zoch and Schober (2018) demonstrated the impact of public child-care on gender norms, and concluded that "family policy reforms may facilitate ideology change not only across cohorts but also during the life course" (p. 1035). It would therefore be interesting to reproduce our original survey methodology in other countries with varied national family policies. This would make it possible to explore more specifically in other societies whether the parenthood norm is also stronger for women than men when using the appropriate survey methodology to activate double standards.

Practice Implications
The findings from this study demonstrate the importance for researchers to develop quantitative survey methodologies that incorporate within-subject designs to test for double standards across the social dimension explored (gender, race or class). Studies that assess endorsement of social norms in a personal background context where the conditions of activation of double standards are not met may produce misleading conclusions about the state of gender, race or class inequalities. This study illustrates the complexity of investigating double standards as well as their importance. All researchers should be aware of these methodological challenges when exploring gender, race or class inequalities.
More broadly, these results call into question the apparent improvement of markers of gender equality in countries with ostensibly higher gender equality, such as France (United Nations Development Programme, 2019). Beyond the individual level where gender egalitarian values seem relatively well-integrated as exhibited in our results within the personal background context, a strong social context still supports gender inequality around parenthood. It is essential to raise awareness among men and women about societal gender norms and their impact.

Conclusion
In the present study, the social relational context revealed a strong gender double standard in the parenthood norm by activating comparisons between men and women. As observed by others, the comparative nature of the social relational context "moves people from personal attitude to public norm" (Schuman & Ludwig, 1983, p. 115). These findings underscore the need to design studies that consider question order and response context when interpreting the results. The present study revealed that the parenthood norm is still strongly gendered in French society despite advances toward gender equality.