Accounting for Gender in the Sociology of Childhood

The (new) sociology of childhood has contributed much to thinking about childhood as a social construction and recognizing the diversity of childhood(s). However, age and generation are often the primary focus of sources in the field. We argue that attention to gender is rare in the sociology of childhood and, when it is mentioned, is under-theorized. Although there is now a growing amount of feminist research with pre-teen children, there is a lack of discussion between the broad field of feminist studies and the sociology of childhood. In this article, we draw on our own research studies in Brazil and Australia, as well as previous key studies about gender and children, to consider the ways in which gender and age can be conceptualized together, and to demonstrate the significance of gender in children’s lives.


Introduction
The (new) sociology of childhood 1 has contributed much to thinking about children and young people, including understanding childhood as a social construction and recognizing the diversity of childhood(s). This way of thinking about children and childhood has been significant in disrupting dominant frameworks of developmental psychology and socialization, to instead conceptualize children as "being" (rather than "becoming") and social actors in their own right. While diversity is mentioned, attention to the specific implications of factors such as gender is largely absent from the sociology of childhood. Ferreira (2004) argues that in the process of socially defining childhood, sociologists removed all the heterogeneity of children, ignoring other factors such as gender, ethnicity, and social class. Relatedly, Eriksson (2007) argues that, "[e]ven though gender has been a central dimension for some of the key researchers in the field . . . so far childhood scholars in general cannot be said to engage deeply with gender studies" (p. 62).
Generally, when gender is mentioned in the sociology of childhood, it tends to be fleeting, in binary terms, and/or is not discussed in relation to feminist gender theories. In this sense, gender is often reduced to a list of differences between boys and girls with no further exploration of how these differences are constructed or what they mean to children. Therefore, what is actually missing is a critical standpoint that accounts for the social construction of gender and the influence of gender relations and hierarchies among children.
There appears to be little dialogue between the sociology of childhood and feminist studies. 2 Recent works by key writers in the field of the sociology of childhood, such as James and James (2012) and Alanen (2012), have celebrated and encouraged the interdisciplinarity of studies of children and childhood, mentioning diverse fields of thought: sociology, social anthropology, psychology, history, geography, philosophy, economics, ethics, medicine, social policy, law, pedagogy, art, and literature. However, there is no specific mention of feminist studies as a current or potentially useful contributor to understandings about children and childhood. Parallels between the sociology of childhood and feminist studies, such as listening to previously unheard voices in research, and the ways in which childhood studies can learn from feminist studies, have been noted, including by Alanen (1994) herself (see also Thorne, 1987). Within feminist studies, children have historically been ignored or not included as research participants (see Alanen, 1994;Oakley, 1994; 580303S GOXXX10.1177/2158244015580303SAGE OpenBartholomaeus and Senkevics  Rosemberg, 2001). This absence likely related to the desire to break the link between women and children, where women and children were viewed as having the same interests and a "natural" connection, and as women tried to separate themselves from maternal responsibilities (Burman & Stacey, 2010). Important early feminist thinking about gender and children appeared in the late 1980s and early 1990s, by writers such as Davies (2003Davies ( /1989, Thorne (1993), and Walkerdine (1990). Since then, there have been an increasing number of studies with children examining gender from feminist perspectives, although it is worthwhile noting that this work has been published far more often in education journals rather than feminist studies or childhood journals. Despite the growth of research about gender and children, many studies do not specifically consider the implications of (socially constructed) age for gender.
In this article, we argue that a consideration of gender drawing on feminist theorizing has much to offer the sociology of childhood. We draw on our own studies, in Australia and Brazil, to provide ways of examining gender and age together and to show how gender is relevant in children's lives. We focus our discussion on two topics-play and sport-which are often discussed in studies with children and relate to our research. Importantly, we situate our research in the contexts in which it was conducted. In some ways, both countries can be viewed as belonging to the "global South," although Australia also has strong connections to the "global North" (for an in-depth discussion of theorizing in relation to the South, see Connell, 2007). Importantly, we are writing from outside the places where the (new) sociology of childhood has been developed and is the strongest, such as in the Nordic countries, the United States, and the United Kingdom. In Australia, children are largely absent from sociology and the national body The Australian Sociological Association (TASA) does not have a thematic group dedicated to childhood (for an overview of children and sociology in Australia, see van Krieken, 2010). Discussions about children and childhood, including in terms of gender, tend to be concentrated in education, where there is a large body of work advocating for feminist-informed teaching and education (for an overview discussion of feminism and education in Australia, see Yates, 2008). Children are also not a focus of sociology in Brazil. While there are some attempts to consolidate a sociology of childhood in Brazil (see Faria & Finco, 2011), as in Australia, most of the research in this regard is undertaken in the education field (Castro & Kosminsky, 2010). At the same time, gender has not been a priority in childhood studies, where it is almost absent in the so-called foundational texts such as Quinteiro (2002).

Absences of Gender in the Sociology of Childhood
In our research conducted in two distinct countries, we noticed independently that gender was largely absent from the sociology of childhood. This is a widespread occurrence in many countries. In a recent special issue of Current Sociology, childhood sociology was mapped in 10 different countries with an article on each country (Australia, Brazil, the United States, the United Kingdom, Finland, France, Germany, Italy, the Netherlands, and Romania). Overall, there is little consideration of gender in these overview articles, and only two mention specific research about children and gender (in the United States and Finland, respectively; Bass, 2010;Strandell, 2010).
There are surprisingly few critiques of the relative absence of gender in the sociology of childhood. An exception is Haavind (2005) who, in a review of several key sociology of childhood books, notes that "[h]ere and there particular chapters hint at observations of gender differences, but this tends to be laid out as a descriptive fact without further reflection on or interpretation of what it means" (p. 150; see also Eriksson, 2007). The lack of attention to gender may relate to the fact that the sociology of childhood is dominated by a small number of influential texts, writers, and concepts that, as Bühler-Niederberger (2010) argues, "produces a global scientific influence which initiates and shapes childhood sociological research" (p. 376). The centrality of age and generation in these writings is also likely to be a factor. In addition, this absence may also relate to the little consideration of children in their own right in feminist studies broadly, as mentioned above. Although we do not expect gender to be a key focus of all sociology of childhood sources, the scarce mention of gender, and the lack of indepth theorizing is surprising. In this section, we provide a brief overview of what has been written about gender in the sociology of childhood, focusing on several key writers and texts in the field.
In the sociology of childhood, gender appears to be discussed most when bodies and/or play are the focus of analysis. For example, Prout's (2000) edited book The Body, Childhood and Society mentions gender more than other key collections in the sociology of childhood. Gender is discussed to some degree in four of the nine chapters. Most notably, Prendergast (2000) draws on feminist theories about bodies, including the work of Grosz and Gilligan, to discuss girls' growing bodies, focusing on their early knowledge about and experiences of menstruation.
A notable writer in the sociology of childhood, Corsaro, occasionally includes some reference to gender in his work on play, peer culture, and interactions. Although Corsaro (2014) references some key feminist thinkers such as Thorne, Walkerdine, Davies, and Connell, his approach often stresses gender differences in terms of distinct peer cultures, as can be seen in his international study comparing the United States and Italy (Aydt & Corsaro, 2003). Moreover, he focuses on differences between boys' and girls' interactions, rather than considering the constructions behind "being" a boy or a girl or the power relations between genders.
Even though gender is sometimes mentioned in the sociology of childhood, it is largely under-theorized. A key text Theorizing Childhood (James, Jenks, & Prout, 1998) includes some discussion of gender and research about gender and childhood, particularly in relation to the work of Thorne. However, for the most part, the authors are less interested in the gender content and theories in these works and instead tend to focus on the contributions to thinking about childhood broadly. The entry for "Gender" in Key Concepts in Childhood Studies outlines that "gender is a critical component of childhood's diversity" (James & James, 2012, p. 61), where it is one aspect of children's social identity that needs to be considered in relation to age, ethnicity, and class (i.e., broader social structures) in the context of their culture and society. However, although James and James draw on some feminist research, engagement with feminist gender theories is notably absent. Thus, gender is occasionally mentioned in discussions of diversity, but without being expanded on. Another example is James and Prout (1997) who acknowledge "the danger that in collectivizing children into 'childhood' significant differences (of gender, class, ethnicity, disability etc.) between children are underplayed" (p. xiii). However, in these accounts, ultimately age is viewed as more important than other differences.
The view that generation and age are the most important categories to examine, at the expense of all other intersections, is most clear in the work of Qvortrup (see, for example, 2000Qvortrup (see, for example, , 2010. He views attention to gender, class, or race as problematic because they assume similarity among children based on particular categories: [t]here is of course nothing wrong in studying gender or class or race issues among children, but one has to be aware that such choices are by the same token not addressing generational questions, i.e. they may be unwarrantedly assuming similarities among children. In this respect Coles (1967: 322) made an important point when he observed that "in a sense white and Negro children have more in common with each other than with their parents." (Qvortrup, 2000, p. 94, n5) Thus, Qvortrup privileges age over any other category, downplaying, for example, shared experiences by children and adults, including in terms of structural inequality. Although many writers in the sociology of childhood view age and generation as the most important category of analysis when studying children, Qvortrup is explicit in his rejection of the need to consider other categories. Indeed, he views them as a threat to the perspective of childhood as a structural element present in every society (Qvortrup, 2010).
The attention to diversity and multiple social structures that Qvortrup so fears is what interests us most. We view age as a significant category in children's lives but not the only category and not the most important category in every context. Even though our focus in this article is on the insufficient attention paid to gender in the sociology of childhood, we acknowledge that this is part of a larger lack of consideration of how age intersects with other forms of diversity too, such as ethnicity, race, sexuality, class, and so on. Rather than considering these as separate categories that can be "added in," we view them as intersecting, and creating contextually specific experiences for children. The rest of the article focuses on feminist-informed research about gender and children, including our own, showing how these studies contribute to thinking about the interweaving of gender and age.

Research About Gender and Children
There is now an increasing amount of feminist-informed research about gender and children that sits largely apart from the sociology of childhood. Broadly, this research examines gender as a social construction. It has been influenced particularly by the work of Connell (2000Connell ( , 2005 and by post-structuralist theories. Connell's theories about the "gender order" and gender hierarchies, particularly in terms of multiple and hierarchical forms of masculinities (see, for example, Connell, 2000Connell, , 2005, have been influential in examining gender relations and patterns among children (as well as teenagers and adults). Her concept of hegemonic masculinity, which denotes the most culturally exalted form and produces complicity and upholds the "gender order," has been frequently used in discussions of young masculinities. Feminist post-structuralist accounts of gender are often used in combination with Connell's theories in research about gender and children. Post-structuralist approaches, such as the work of Foucault and Butler, emphasize the fluidity of subjects and view power as produced through discourse. Although there is diversity among writers drawing on these theories for research about gender and children, Connell's gender theories and feminist post-structuralist approaches have been most frequently drawn on and therefore appear useful for theorizing gender in the sociology of childhood.
In this section, we draw on our own research in Brazil and Australia, focusing on the ways in which we use gender theories in our research, and highlighting the contributions they make to understanding gender and age. Although we are aware of the difficulties of discussing two different studies together, we believe that they are useful to explore how sociological studies of gender among children can be undertaken in different countries. To focus the scope of this article, we offer a consideration of two topics: play and sport. Play is a topic that has received much attention in work about childhood and sport is a key theme discussed in studies about masculinities and boys. To provide a context for our own research, we start both sections with a brief discussion of some of the existing examples of research studies that have considered the interweaving of gender and age in childhood.

Feminist Research About Play, Gender, and Children
Play in its various forms is often a focus of research with children and, therefore, is an important topic to examine in terms of what has been written about gender. Research about play, children, and gender has been influenced by and built on feminist-informed gender theories, including the view that gender is fluid and contextual, that ideas about gender can influence and restrict children's play, and that there are power relations between and within genders.
How gender, play, and children are conceptualized has been greatly influenced by Thorne's (1993) foundational book Gender Play, which draws on ethnographic research in the United States in 1976 to 1977 and 1980. Thorne's work is influenced by postmodern feminist theory (specifically Flax, 1987, Butler, 1990, and Scott 1988 and the work of Connell. Thorne demonstrates that play can be fun but it is not trivial-it is important to understanding gender constructions among children. Thorne (1993) developed the influential concept of "borderwork" to examine "interaction acrossyet interaction based on and even strengthening-gender boundaries" (p. 64). She examines the different kinds of borderwork that take place in elementary schools-contests between boys and girls, chasing/kiss chasing, and ideas of gender pollution-showing how these work to strengthen notions of "boys" and "girls" as opposites. Thorne shows that both age and gender are important in the lives of children at school, along with other factors (such as class and ethnicity), and that context is central to the significance of gender and age.
Davies also offers a sophisticated consideration of the intersection of gender and age. Her work has some similarities to Thorne's in that she draws both on Connell and poststructuralism and moves away from understanding gender as a binary, although Davies offers a more thorough-going poststructuralist account. Davies's (2003Davies's ( /1989) research with preschool children in Australia, documented in Frogs and Snails and Feminist Tales, shows how gender affected children's lives in a number of ways, from how they could play in the "home corner" to awareness of male power to the way in which dress-up clothes were viewed by children as infused with multiple meanings and possibilities. She draws on the concepts of the "male-female binary" and "category-maintenance work," arguing that "category-maintenance work" occurs around gender boundaries where "deviants" are policed and gender categories are maintained regardless of individual deviations (Davies, 2003(Davies, /1989). Davies's post-structuralist account of theorizing gender means that she views gender as an ongoing construction, never fixed and never unitary, where individuals are positioned in different ways in different contexts. Ferreira (2004) offers an example from a Portuguese context in her book A gente gosta é de brincar com os outros meninos (We like to play with the other boys), which draws on an ethnography in a public kindergarten with 3-to 6-yearold children. Drawing particularly on Scott's post-structuralist approach to gender, Thorne's ethnography, and Connell's conceptualization of masculinities, Ferreira pays careful attention to how gender is actively reproduced and/or transformed by children throughout their play and interactions.
She argues that children negotiate plural, unstable, and contradictory "gender identities," and thus no child simply embodies a coherent model of being a boy or a girl. Paying attention to children's participation in this process, the author adopts the notion of "dealing with" a certain "model," "ideal" or "discourse." In addition, Ferreira (2004) acknowledges the coexistence of multiple masculinities and femininities as well as "their ludic and/or strategic usage by the same actor, according to the circumstances" (p. 358; authors' translation).
Much of the research about gender and play is undertaken in education settings, which are seen as important places for the construction of childhood. However, many other places are relevant sites where children construct gender as they play, learn, and interact, such as their local neighborhoods. With this in mind, the second author's research in Brazil provides an example of how play, access to neighborhood spaces, and domestic work intersect and relate to gender.

Playing Outdoors, Working Indoors: A Consideration From Brazil
Conducted in Brazil, the second author's research draws on a qualitative study examining gendered family practices among children from working-class backgrounds in São Paulo city. The data were collected at one public school, where children gave their viewpoints about their school, parents, siblings, and home. These results were useful to understanding how gender affects children's schooling achievements considering, as a starting point, that girls overall have better outcomes than boys in education in Brazil (Carvalho, 2009;Rosemberg & Madsen, 2011). A key finding of the research was the leisure-housework balance, where the possibility to play on the street compared with the constraints of working at home was experienced differently depending on the gender of the children.
The research was undertaken in 2012, in a co-educational public school with students from urban working-class communities (in this context, children came from families who worked in low-level and often informal jobs). One Year 3 class participated in the research, involving 25 students who interacted with the researcher around twice a week for one semester. After a couple of weeks informally talking and playing with the students, six individual and seven paired interviews were conducted with 20 children (12 girls and 8 boys), aged mostly 8 or 9 years. During the interviews, children were asked about their routines at home and the range of activities they engaged in once they left school in the afternoon. Connell's (2005) concept of masculinities and her emphasis on social practices were adopted to examine what children did in their daily routines, how these activities related to gender, and the potential effects of these on the schooling process among girls and boys.
A sharp difference in the leisure time, range of play activities, and access to the street was found between boys and girls (as also found by Fiaes, Marques, Cotrim, & Bichara, 2010;Silva, Pontes, Silva, Magalhães, & Bichara, 2006). While the street was a site that boys could often access, meet friends, and have fun through the practice of play and sports, risks concerning the public space were often pointed out by girls. Fear of being kidnapped, meeting drunk people, or even catching influenza were evoked to compose a list of dangers that were expressed by girls and could be seen as encouraged by gendered parental concerns that related to distinct rules and expectations among their offspring. None of the girls felt that they were free to play on the street as much as they wanted. The public space itself was perceived as dangerous for girls, while for a lot of boys those risks did not appear to be a limit in their daily routines, even though they lived in the same communities as their female classmates. Thus, masculinities were constructed taking as a reference point the public space and the range of games and activities that boys could play after the school. Playing on the street gains a special importance in working-class children's lives in Brazil because leisure opportunities are scarce. No public gardens or parks characterize low-income communities in São Paulo, nor do these houses have yards to play in. Without entering the public space, there remain only the small spaces of houses to play in, where children undertook activities such as playing with dolls, watching television, playing videogames, or using the computer. That is why having access to the outdoors (playing football on the street, having fun on the sidewalks, or interacting in any empty land) is so relevant for children who were allowed to dedicate part of the day to play there.
Alongside perceptions of risk, girls' participation in housework diminished their access to leisure time, and to play on the streets and elsewhere (see also, for example, Valentine, 2004). This process may have influenced the construction of femininities defined by responsibilities and discipline that were rare features among their brothers and male classmates who spent most of the day playing. This fed back into the school success among girls that, according to Carvalho, Senkevics, and Loges (2014), seems to rise from within the practices that constrained girls' access to leisure and sociability. In other words, girls' limitations to playing meant that they spent more time on school work, had more of an interest in going daily to school, and had a more positive relationship with this institution than boys.
In addition, two contradictory processes were evident in children's interviews. On one hand, girls were critical of their brothers' low level of engagement in housework, reporting situations when boys had more opportunities to play in or outside the house. Such inequality was not readily accepted and, though girls continued to do housework, they also criticized boys' benefits. On the other hand, participating in the gendered division of domestic work was in a certain way naturalized among girls. For most girls, doing housework was not a kind of "help" or "contribution," but a responsibility and a duty. The sharing of responsibilities was implied among sisters, which did not occur between sisters and brothers. Moreover, only one boy was really engaged in domestic work. Interestingly, this boy had few leisure opportunities, no siblings of a similar age to play with, and almost no access to public space. This research suggests that, even within the same family, children may experience distinct childhoods due to gendered expectations, according to which housework is part of a gendered routine that mostly relies on girls' engagement. In contrast, having the whole day to play, especially outside home, is a male privilege.

Feminist Research About Sport, Gender, and Children
Commonly viewed as a form of play, sport is often a key focus of feminist-informed research about boys and masculinities in primary school, especially in Anglophone countries. Sport tends to be written about in terms of school settings rather than early childhood settings, highlighting the importance of age and context. Studies with primary school boys frequently draw on Connell's (2000Connell's ( , 2005 theoretical framework of a hierarchy of multiple masculinities, particularly her concept of hegemonic masculinity (see, for example, Bhana, 2008;Renold, 1997;Swain, 2006). Such studies often examine sport in relation to privileged and hierarchical masculinities as well as hierarchical relations between masculinities and femininities.
The role of peer group settings is particularly important for boys' constructions and understandings of masculinities, including in relation to sport. Keddie's (2003) research with five male friends aged 6 to 8 years in an Australian primary school shows the importance of gender in the lives of these boys by examining their investments in (Australian Rules) football and how this links to gendered, heterosexist, and homophobic understandings of masculinity (drawing on the work of Connell, among others). Keddie draws on feminist post-structuralist theorizing, particularly the work of Davies, to examine dominant understandings of gender. She uses post-structuralist theorizing to be able to examine how social power was exercised and to reconsider "dominant and dominating ways of being" (Keddie, 2003, p. 75). In addition, by closely examining small-group peer culture, Keddie's research illuminates how gender was collectively constructed and understood, including the discrepancies and challenges to dominant understandings of masculinity.
Sport has also been examined in terms of the ways in which gender interweaves with race and class in childhood. Bhana (2008) analyzes the significance of sport among 8and 9-year-old boys in two distinct primary schools in South Africa. She discusses the multiple hierarchies at work, where sport was key to the privileging of some boys over "weaker" boys and girls. In addition, some sports were viewed as more "masculine" than others, which was notably contested among the White and Black boys. Bhana utilizes Connell's concept of hegemonic masculinity to examine gender hierarchies in her research, which she analyzes particularly in relation to race and class. She considers boys' understandings and constructions of masculinities and sport as being influenced by the contexts of the schools, the areas in which the students lived, and broader aspects in South Africa, such as the history of rugby and middle-class White men, and soccer and working-class Black men.
Research has found that despite their young age, boys (and girls) draw on broader discourses about gender and sport, including those in the media and professional sport (see also, for example, Keddie, 2003), to construct masculinities in particular ways and create gender hierarchies. Even though boys do not have access to these adult masculinities, they can still act them out in some ways. Bhana (2008) notes that regardless of the physical bodies of boys, it is important to recognize that strong, sporting adult bodies help shape discourses around gender for boys. Lodge (2005) similarly argues that "[e]ngagement in football and aptitude in the game allows young boys to participate in a game that is also a high status adult male activity" (p. 184). The intersection of gender and age in terms of sport is discussed in more depth below in relation to the first author's research in Australia.

Sport and Gender Hierarchies: A Consideration From Australia
The first author's research examined constructions and understandings of gender in two Australian primary schools. The research focused on the intersection between gender and age, and considered the usefulness of implicitly adult gender theories, particularly those about masculinities, to understand gender among primary school-age students. A key finding of the research was the importance of sport in constructing a privileged masculinity among boys, as well as upholding a distinction between boys and girls, where boys were viewed as superior to girls (for an earlier discussion, see Bartholomaeus, 2011). This research and the previous studies discussed above demonstrate the complex gender and age dimensions of sport in childhood, where sport is more than a social or fitness activity.
Research was conducted in two co-educational primary schools in metropolitan Adelaide, South Australia, in 2009. St. Catherine's Primary was a small Catholic school with over half of the students from Anglo/White backgrounds. Socrates Primary was a Greek Orthodox school, comprising mainly students from Greek and/or Cypriot backgrounds. Both schools were largely middle class, with students from Socrates Primary more likely to be from privileged backgrounds than those from St Catherine's Primary. Two classes from two age groups participated at both schools, involving a total of 95 students aged 6 to 7 and 11 to 13 years. Students participated in a number of verbal and written activities in groups and individually across five sessions that provided them with different avenues to reflect on their ideas about being girls and boys (for full details, see Bartholomaeus, 2012). Class teachers and interested parents were interviewed about how they thought gender was understood by the students.
In this research, sport and physical bodies were drawn on to construct a legitimating masculinity that influenced many boys and was also supported by a number of girls. The pervasiveness of this idea was highlighted by the fact that the research was conducted solely in the classroom, which meant that the author did not interact with the students when they were playing sport. The author reconsidered Connell's (2000Connell's ( , 2005 multiple masculinities framework in light of the data, noting where her theories were useful, and where the findings could not be easily interpreted into the patterns she writes about. A consideration of the impact of young age on gender indicated a need to understand gender as more fluid than Connell's theorizing allows for. Applying Foucault's notion of discourse was most beneficial in theorizing this fluidity, yet many of Connell's theories were still useful in theoretically framing the findings. The author followed the lead of others who have drawn on the notion of a discourse of hegemonic masculinity (Beasley, 2008) to theorize that which is most influential in defining what is most "masculine" in any given setting, and that ensures men's (as a group) authority over women (as a group). As Beasley (2008) writes, "hegemonic masculinity [may be viewed] as a political mechanism-as a discursive ideal mobilizing legitimation" (p. 100). This concept enables an examination of the fluidity of other practices that work alongside or present challenges to a discourse of hegemonic masculinity, including fluidity relating to age and context (Bartholomaeus, 2013).
Therefore, the privileging of sport for constructing masculinities could be understood as a discourse of hegemonic masculinity. Sporting masculinities were the most privileged when considering different masculinities in relation to each other; sport was constructed as something that was important to being a boy; and sport was often constructed as being for boys and not for girls. The cultural exaltation of sporting masculinities was evident when students were asked to rank eight famous faces from most "manly" to least "manly," giving reasons for their choices. Athletes were commonly viewed by the students as the most "manly," above other men such as actors and the then Australian Prime Minister (Kevin Rudd). Athletes were not privileged in the same way in a related activity ranking female famous faces from most "womanly" to least "womanly." The privileging of sporting masculinities was also evident when students wrote about people they looked up to, where it was clear that boys looked up to only male athletes, and that these represented adult masculinities. In contrast, none of the girls wrote about looking up to any athletes.
Sport was also constructed as being for boys by claiming that it was not for girls. This included the view that girls do not play sport or are less skilled at it than boys (see also, for example, Renold, 1997;Swain, 2006). At the same time, some girls, as well as some boys who did not like or engage in sport, also equated sport with boys. Generally, girls' involvement in sport was mentioned by girls only when provoked or challenged by boys, such as when boys made claims to be more skilful at sport than girls. Importantly, alongside the construction of boy/girl divisions in terms of sport, hierarchies of sports were used to create a discourse of hegemonic masculinity (see also Bhana, 2008;Keddie, 2003;Swain, 2006), with soccer being the most privileged for boys at both schools.
Key broader influences on constructing sport as important for masculinity included the media, schools and teachers, and parents. For example, the lack of women's sports shown in the (Australian) media was drawn on by a 6-year-old boy in particular to ascertain that "only men and boys play soccer." Another example of a broader influence came from the mother of a Year 6 boy who suggested that expectations in terms of gender coming from her husband involved "push[ing]" their two sons into sport, which was not the same for their daughter.
This research supports the contention that, in Australia at least, it appears that a discourse of hegemonic masculinity begins with sport. Importantly, due to their young age, boys are unlikely to have access to other factors often related to hegemonic masculinity, such as work, sexuality, and/or fatherhood (Connell, 1983). However, students often drew on adult men in their understandings of privileged masculinity in terms of sport and to enhance the gendered power relations in their own lives. The young age of the students was a barrier to accessing a broader discourse of hegemonic masculinity, even to those boys who embraced sports and associated aspects of physicality. This research shows how the complex interweaving between gender and age plays out in relation to sport.

Conclusion
In this article, we have argued that though gender is not a key concern in the sociology of childhood, it is an important factor in children's lives. When gender is included in this field, it tends to be under-theorized, where gender is often reduced to a list of differences between boys and girls or it is simply mentioned as one factor relating to diversity. Drawing on a feminist analysis of gender together with a sociological approach to the study of children and childhood helps to account for the varied positions that children occupy in society and their experiences, including in relation to gender.
By drawing on our own research in Brazil and Australia, as well as previous research, we explored how gender was taken up and constructed in children's lives and the ways in which this can be theorized. We examined the complex interweaving in Brazil of play and other factors such as responsibilities in housework and risks concerning public space, with specific attention to how gender influenced how this was experienced by children. Although children's experiences in terms of access to leisure and constraints relating to housework were often divided by gender, this can be seen as a construction often put in place by parental and broader societal views of risk and responsibilities in terms of girls. Indeed, some of the girls expressed critical standpoints against the imposed gender dualism. The research in Australia showed how gender and age interweave to produce particular privileged gender discourses, where masculinities are often constructed via sport and as distinct from femininities. Importantly, a discourse of hegemonic masculinity based on sport was something that many boys aspired to but did not necessarily have access to. Thus, power relations between different practices of masculinities were evident as well as between masculinities and femininities.
Despite the differences between our two studies, some commonalities in our approach should be highlighted. Broadly, we both argue for the importance of examining children's experiences of gender and how gender relations play out between boys and girls as well as among boys and among girls. Furthermore, a consideration of the power relations inherent in interactions and structures is necessary, including to better understand children's pleasures and felt restrictions in relation to gender. We feel that it is important to consider how children engage with, construct, take up, and challenge dominant gender discourses as well as to examine how gender matters in children's lives, including when it is important and when it is less important, and how age influences this. In particular, we want to go beyond an examination of gender binaries to explore the complexity of gender in children's lives. This is not to deny that gender can work as a binary in children's lives, including how they may explicitly talk about it, such as constructing sport as for boys and not for girls, and how it restricts their everyday activities, such as girls not being allowed to play on the street when boys are. Going beyond identifying binaries and viewing them as taken for granted, it is important to examine the social and cultural mechanisms keeping this binary working, and how children experience it.
More broadly, we feel that it is important to reconsider the concept of "child" in the sociology of childhood. We would like to ask how much can be inferred from the experiences of children if they are devoid of other social and structural factors and contexts. We agree with the central tenet of the sociology of childhood that "childhood" is both a social construction and a structural category that is important in children's lives. However, the tendency to view age as the most important category in all contexts, places, and times ignores children's experiences of other factors, and how these factors intersect with age to produce specific experiences and positions. Although we have focused on gender in this article, there is also clearly a need to extend conversations about diversity in the sociology of childhood. This involves paying attention to other social categories such as ethnicity, race, sexuality, and class. and considering how these intersect with age (and gender). Exploring the ways in which age interweaves with gender goes some way toward enhancing the discussion about accounting for diversity among children, which is mentioned in the sociology of childhood but rarely examined in depth.