The ‘mighty girl’ effect: does parenting daughters alter attitudes towards gender norms?

Mireia Borrell-Porta Joan Costa-Font Julia Philipp Oxford Economic Papers, gpy063, https://doi.org/10.1093/oep/gpy063 Published: 14 December 2018 Abstract We study the effect of parenting daughters on attitudes towards gender norms in the UK; specifically, attitudes towards the traditional male breadwinner norm in which it is the husband’s role to work and the wife’s to stay at home. We find robust evidence that rearing daughters decreases fathers’ likelihood to hold traditional attitudes. This result is driven by fathers of school-aged daughters, for whom the effects are robust to the inclusion of individual fixed effects. Our estimates suggest that fathers’ probability to support traditional gender norms declines by approximately 3%age points (8%) when parenting primary school-aged daughters and by 4%age points (11%) when parenting secondary school-aged daughters. The effect on mothers’ attitudes is generally not statistically significant. These findings are consistent with exposure and identity theories. We conclude that gender norm attitudes are not stable throughout the life-course and can significantly be shaped by adulthood experiences. Issue Section: Original Article 1. Introduction In recent decades, concerns about gender equality have been increasingly prominent in both the political and the social spheres, prompting governments to embark on the task of alleviating gender differences inside and outside the labour market. Nevertheless, progress towards achieving gender equality appears to have gradually slowed down (Eagly and Wood, 2012; Gender Equality Index, 2017). Against this background, a growing body of research has established the importance of traditional gender norms in explaining the persistence of gender inequalities in wages (Burda et al., 2007), in labour force participation (Fernández et al., 2004; Fortin, 2005; Fernández and Fogli, 2009; Farre and Vella, 2013; Johnston et al. 2014), and in the division of domestic work (DeMaris and Longmore, 1996; Greenstein, 1996, and see Davis and Greenstein 2009 for a review). However, there is limited evidence on how susceptible to change such norms are. This paper addresses this question. Changing individual attitudes towards societal gender norms may be critical for further progress towards a less gendered division of work and towards gender equality more generally, since it may legitimize a wider range of social roles for both men and women (Eagly and Wood, 2012). So far, the literature has focused mostly on long-term changes in norms across cohorts (Baxter et al., 2015), although research on individual changes in attitudes towards gender norms across the life cycle is gradually increasing. This latter approach has studied the role played by the family environment, including that of marriage, parenthood, and women’s labour patterns inside and outside the household (see Clarkberg, 2002; Corrigall and Konrad, 2007; Cunningham, 2008; Schober and Scott, 2012; Baxter et al., 2015). Our paper contributes to this literature by analysing one life course event that has received limited attention so far, namely the effect of parenting daughters. Using a British nationally representative longitudinal survey spanning two decades, we examine whether rearing daughters changes parental attitudes towards gender norms, and more specifically, attitudes towards the traditional male breadwinner norm in which it is the husband’s role to work and the wife’s role to stay at home. Given that a child’s gender1 cannot be anticipated, we assume that rearing a daughter—as opposed to a son—is an approximately random event (Washington, 2008).2 In examining the individual change in attitudes3 towards gender norms, we borrow the definition of gender norms from Pearse and Connell (2016), who define them as ‘collective definitions of socially approved conduct in relation to groups constituted in the gender order—mainly distinctions between men and women’. Hence, norms are defined as ‘features of a collective life’ (p.34) that signal to other members of a group or society how they should behave’ (Schwartz, 2012, p.16, emphasis added), and closely follows definitions in other social science disciplines.4 We find evidence that parenting daughters decreases fathers’ likelihood to agree with a traditional male breadwinner norm. This is especially the case for fathers of school-aged daughters, for whom the effects are robust to a number of alternative specifications, and in particular the inclusion of individual fixed effects (FE) that control for time-invariant unobserved heterogeneity. Our FE estimates suggest that fathers’ probability to support traditional gender norms declines by approximately three percentage points (8% change) when parenting primary school-aged daughters and by four percentage points (11% change) when parenting secondary school-aged daughters. In contrast, the effect on mothers’ attitudes is smaller and generally not statistically significant. While it is not possible to discern the exact mechanisms through which daughters affect parental attitudes, the heterogeneity of results between fathers and mothers combined with the finding that attitudinal change occurs when daughters reach school age is in line with theories of exposure as well as with identity theories. Furthermore, given that attitudes towards gender norms are shaped by experiences during adulthood, our results provide evidence of intra-cohort change in attitudes. Consistent with our findings on attitudinal change, we find that parenting school-age daughters is also associated with a lower likelihood that couples follow a traditional gender division of work. To the best of our knowledge, this is the first paper to explore the impact of child gender across daughters’ ages on individual changes in attitudes towards gender norms. This is important because our findings suggest that it is when daughters are of school-age—and not before—that fathers’ attitudes become less traditional, thus coinciding with the period in which children experience a stronger social pressure to conform to gender norms (Lane et al., 2017). The paper also contributes to expanding the evidence beyond the USA, being the first paper to explore the impact of the gender of the child on attitudes towards gender norms in the UK. Finally, and unlike previous studies, we draw on data that covers very recent years—up to 2012—which is important given the large changes in patterns of gender inequalities during recent decades. The structure of the paper is as follows. The next section reviews the relevant literature, and Section 3 describes the data and empirical strategy. Section 4 contains the main results, Section five robustness checks, and a final section concludes. 2. Related literature 2.1 On the malleability of attitudes towards gender norms There are two main approaches in social science on the evolution of attitudes towards social norms, including gender norms. One approach suggests that attitudes are formed before reaching adulthood and remain stable thereafter.5 Societal change in norms then occurs through processes of cohort succession, when older cohorts are replaced by younger ones who systematically differ in its social and historical early years’ experiences (Mannheim, 1952; Brooks and Bolzendahl, 2004). An alternative approach—the one embraced by this paper—questions the stability of norms and embraces the viewpoint that attitudes can change over the life course, either due to social structural changes, or due to changes in individual circumstances (Brooks and Bolzendahl, 2004; Hogg and Vaughan, 2008; Baxter et al., 2015).6 Empirical evidence concerning these two approaches remains inconclusive. Two early papers analysing US data from the 1970s and the 1980s point in different directions, one suggesting that attitudes towards familial roles in the USA occur mainly within cohorts (Mason and Lu, 1988) and another providing evidence for cohort replacement-based explanations (Wilkie, 1993). Further analyses with data from the 1990s and early 2000s have not resolved the debate. While some research supports cohort replacement theories (Brewster and Padavic, 2000), there is also evidence which confirms the importance of intra-cohort change (Danigelis et al., 2007). Another paper finds that while cohort replacement theories have a strong explanatory power, ideological learning during adulthood may mediate a large part of the cohort replacement effect (Brooks and Bolzendahl, 2004). Attention on intra-cohort change has recently shifted the focus of research towards the potential factors underpinning change, with a particular emphasis on family environment. To this purpose, longitudinal data has increasingly been used to study the impact on attitudes towards gender norms of women’s decision to work (Cunningham, 2008), parenthood (Evertsson, 2013; Baxter et al., 2015), the interaction between work and childbirth (Berrington et al., 2008; Schober and Scott, 2012), and marriage and cohabitation (Moors, 2003; Corrigall and Konrad, 2007). Nonetheless, the gender of the child has received limited attention. In what follows, we will focus on the specific effect of child gender. 2.2 On the relevance of child gen