“Any advice is welcome isn’t it?”: neoliberal parenting education, local mothering cultures, and social class

. Geographers have shown considerable interest in neoliberal educational restructuring as states across the Global North have sought to respond to the challenges of economic change through the development of a skilled population. Existing research provides a wide-ranging analysis of the ways neoliberal states seek to shape individual citizens through their own learning. Greater attention now needs to be paid to new and developing ways in which they seek to influence the context in which future citizen-workers are raised. This paper focuses on parenting education which is growing across OECD countries. Social science critiques suggest that parenting classes are part of a professionalisation of parenting which has sought to impose middle-class mores on working-class parents, at the same time as parenting has been unwarrantedly cast as a context-free skill. This paper uses quantitative and qualitative data to explore the attitudes of parents of different social class positions to parenting education, tracing the ways these emerge in and through particular sociospatial contexts. The paper reveals the importance of local class-based cultures of mothering in influencing both the attitudes of individual mothers to parenting classes, and the success of neoliberal policy implementation in diverse socioeconomic neighbourhoods. In conclusion the paper emphasises: the importance of geographical research into newly emerging forms of education; the value of engaging with the subjects of neoliberal education policy because their attitudes influence its implementation in practice; and the need to set educational provision in its wider geographical context, as this can shape the success of policies delivered in and through educational institutions.


Introduction
Education has risen up the political agenda as welfare states in the Global North have sought to respond to the challenges of economic restructuring through the development of a skilled population (Jenson and Saint-Martin, 2006). This increased prominence in the political arena has been matched by a growth in geographical writing, with the field developing especially quickly in the 21st century (Hanson Thiem, 2009;Holloway et al, 2010;Power and Malmberg, 2008). The result has been a wide-ranging analysis of the education provided for, and experienced by, diverse individuals, be it in childcare, schools, universities, or alternative learning spaces (Kraftl, 2013;Purcell, 2011;Smith and Barker, 2000;Waters, 2009). Neoliberal states are now seeking to shape the development of future citizen-workers in new and expanded ways (Holloway and Pimlott-Wilson, 2012;Wainwright and Marendet, forthcoming) and these developments require geographical attention. In this paper we focus specifically on parenting education, a means through which the state is not simply seeking to shape individual citizens through their own learning, but more broadly shape the context in which future citizen-workers are raised. The paper begins below by tracing the rise of parenting education in the context of 21st-century neoliberal restructuring and examines academic criticism about class biases inherent in the wider professionalisation of parenting of which it is a part. The subsequent section introduces our empirical research with class-differentiated parents of primary-schoolaged children in England. The central sections of the paper explore these parents' attitudes to state provision of parenting education, drawing out differences in the views of those with children in higher, middle, and low-income primary schools, and analysing how their attitudes emerge in the diverse socioeconomic contexts in which they parent. In conclusion, the paper emphasises the importance of examining new forms of education which are emerging with neoliberal restructuring, and the views and opinions of those who are the intended subjects of policy. Geography matters in this endeavour, the paper shows, because educational aspirations and experiences are mediated through sociospatial contexts beyond the walls of individual educational institutions.
Research by Shulruf et al (2009) shows that parenting education and support have been attracting increased attention, and expenditure, across a range of OECD countries. Treating this as a 'universal' trend would overplay the homogeneity of neoliberalism (Larner, 2003), and we therefore explore how this developed in one particular nation. In England, the empirical focus of this paper, the Labour government's time in office (1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010) was characterised by a form of roll-out neoliberalism popularly summarised at the time as being about 'rights and responsibilities'. On the one hand, Labour sought to reform the welfare state by pursuing a workfare agenda, but on the other hand they acted to renew it by increasing state intervention in child and family policy as this was seen to promote social justice (Holloway and Pimlott-Wilson, 2012). This included a marked expansion of the state's role in parenting support as parents were envisaged as crucial actors in shaping children's current and future social in/exclusion. The Every Child Matters Green Paper (HM Treasury, 2003), for example, identified parenting as vital in shaping both positive and negative outcomes for children, a belief further evidenced in Alan Johnson's (then Secretary of State for Education) forward to Every Parent Matters (DfES, 2007a, page 3) where he stated " [p]arents and the home environment they create are the single most important factor in shaping their children's wellbeing, achievements and prospects." This belief led New Labour to overcome previous reticence about state involvement in parenting and to use parenting interventions to tackle both social exclusion and antisocial behaviour (James, 2009), tasking parents with inculcating both educational aspirations that would facilitate children's social inclusion as adult citizen-workers, and codes of conduct that would ensure current and future social stability (Holloway and Pimlott-Wilson, 2011). In this way, the roll-out neoliberalism of New Labour led to the development of a multifaceted parenting support agenda which spread across a range of Government Departments, including that with responsibility for education (Utting, 2009).
A Conservative-Liberal Democrat Coalition replaced Labour after the May 2010 general election. This has prompted a significant change in the nature of neoliberalism in England (Featherstone et al, 2012) and has seen the new government seeking to roll back the state through swingeing cuts to public spending as part of a self-styled age of austerity. The Coalition's approach to parenting education has been inconsistent. On the one hand, their small-state philosophy comes through in rhetoric which cautions that "[p]romoting good parenting is not primarily a job for the Government" (HM Government, 2011, page 38) and funding cuts to local government have caused cuts in services. On the other hand, they too argue that they "want access to parenting advice and support once a child is born to be considered the norm" (HM Government, 2011, page 38) and the Prime Minister has personally championed the universal availability of parenting classes for parents with younger children, rejecting claims that this promotes a 'nanny state' (BBC, 2012). This service is currently being trialled in pilot areas where classes are being made available to all through a mixed economy of state funding and independent provision (DfE, 2012).
These developments in parenting education, and wider child and family policy, have led social scientists to argue that there has been a professionalisation of parenting, in which parenting has been reframed from a "personal family relationship" to "a technical exercisesomething that you can either get right or wrong" (Gillies, 2010, page 44). The fact that parenting is no longer taken for granted and is instead regarded as a skill which you can teach and learn is, Klett-Davies (2010) argues, the basis of a new parenting culture in England. This professionalisation of parenting has attracted considerable academic attention in the broader social sciences, and two criticisms of the assumptions that underpin it are of interest to us here. Firstly, there is concern that the desire to promote good parenting through various facets of child and family policy has not been class neutral, as policy has been shaped by middleclass values, with working-class parents being encouraged to behave in middle-class ways (James, 2009;Klett-Davies, 2010). Wider writing on the neoliberal state in geography might lead us to expect this, as Raco (2009, page 443), for example, argues that New Labour's term in power saw "the normalisation and mainstreaming of practices and ways of thinking that, in fact, reflect a narrow form of [middle-]class-infused consciousness." There is, however, a need for precision in our analysis in this case, as not all forms of middle-class parenting are being given legitimacy. Permissive and liberal parenting, for example, are not endorsed (Cruddas, 2010); rather a model described in both Labour and Coalition policy documents as 'warm and authoritative' parenting is being normalised, and this emphasises the importance of parental responsiveness alongside clear boundary setting (DfES, 2007a, page 28;DfE and DoH, 2011, page 35).
Secondly, the idea that parenting can be taught and learnt has been subject to criticism because it depends upon the assumption that parenting is a context-free skill. The emphasis on improving individual parents' abilities, regardless of the circumstances in which they care, reflects a wider change in the neoliberal state where the policy focus is no longer society, rather the aim is to change individual actors so that they might better perform their responsibilities as future citizen-workers (Raco, 2009). It is evident both in Labour pronouncements on parents' individual responsibility for outcomes for their children (quoted above), and more recently in Coalition policy discourse: " Mothers and fathers are their children's first and most important educators. … What happens in this home environment has more influence on future achievement than innate ability, material circumstances or the quality of pre-school and school provision" (DfE and DoH, 2011, page 36). The notion that parenting is a context-free skill is rejected by social scientists who argue that working-class and middle-class parents face different challenges and that their parenting practices are necessarily shaped both for and by these social contexts (Edwards, 2010;Reay, 2010). As Gillies (2010, page 59) argues: " [t]he notion that childrearing is a 'skill' that can be practised independently from the social context bears little analysis. This formulaic 'parenting by numbers' approach fails to engage with the reality of life for many parents and children." These two critiques-that policies which have promoted the professionalisation of parenting have been shaped by middle-class mores, at the same time as parenting has unwarrantedly been cast as a context-free skill-inform our agenda in this paper. Our first aim is to examine the attitudes of parents of different social class positions to the provision of parenting education. Geographic research demonstrates that neoliberal policy is not simply implemented with inevitable consequences for subject formation (MacLeavy, 2008) and it is therefore necessary to consider the attitudes and experiences of those who are the intended subjects of education policy: in this case, parents (Holloway et al, 2010). Parents are not a homogenous group, and consequently we explore how their views on the provision of parenting education vary with class background. Our second aim, in doing this, is to explore the importance of sociospatial context in shaping parents' attitudes to this service. Social scientists have demonstrated that parenting cannot be seen as a context-free skill: class matters as middle-class and working-class parents face different challenges and have differing resources with which to respond to these. In highlighting the importance of social context, however, they have not paid sufficient attention to how this is experienced in and through different places (Vincent et al, 2010). In this paper we seek to spatialise this notion of social context, exploring how class cultures are made in and through a variety of different neighbourhood spaces. These intersecting aims are taken forward through an empirical study based in England which we introduce below.

Methodology
Our focus in this paper is on attitudes to the provision of parenting education through primary schools in England. Provision of parenting education through primary schools was stimulated in the first decade of the 21st century by the Extended Services policy introduced by Labour which asked primary schools to provide or sign-post parenting support, including measures such as parenting classes, subject to their own assessments of local need (DfES, 2007b). The policy continues under the current Coalition government, although funding cuts to local authorities and the removal of ring-fencing around funding to schools have caused significant difficulties. This paper draws on results from an empirical study into the implementation of this policy in a provincial local authority in England which for reasons of confidentiality we refer to by the pseudonym Hortonshire. Hortonshire was chosen as the location for the study as it was well advanced in the implementation of Extended Services. Moreover, the geography of Hortonshire meant that it contained schools serving children from different class backgrounds, whilst overall the authority roughly conformed to national averages in terms of the number of children receiving free school meals (DCSF, 2009). Children were living in a mixture of large urban, smaller urban, and rural communities. These provincial settlements were less ethnically diverse than England as a whole: over 95% of residents were White British, compared with the national average of 87% (ONS, 2005).
The study used a mix of quantitative and qualitative methods. We began with a questionnaire survey of all parents with children in year 2 (age 6-7) or year 6 (10-11) in twenty-six primary schools asking them about their attitudes to Extended Services, including parenting education and support (n = 722, a response rate > 40% in all areas). In this paper we focus specifically on attitudes to elective parenting classes through primary schools which are designed to promote the 'warm and authoritative parenting' described above, and therefore cover amongst other things strengthening family relationships and managing behaviour. [We discuss other forms of parenting education elsewhere, for example, guidance given to parents on how to help their children learn in the home (Holloway and Pimlott-Wilson, 2013)].
As our research aim was to explore class differences in attitudes to the provision of parenting classes through schools, we chose to contact parents through three categories of primary schools: those with the most economically advantaged intakes; those serving families in average circumstances; and those drawing on the most financially impoverished communities. Schools were classified on the basis of their free school meal (FSM) rate as this is the key proxy for class in the English school context where schools do not hold information on parental income. Access to FSMs depends upon parental eligibility for means-tested benefits; thus we can tell that schools with lower FSM rates have fewer children living in financially impoverished circumstances on their roll than those with higher FSM rates. Schools with an FSM rate below 2.5% were classified as higher income (HI); those with a FSM levels around the local authority and national average (percentages in the teens) were classified as middle income (MI); and those with FSM eligibility over 30% were classified as low income (LI), as the Office for National Statistics (ONS) defines such schools as deprived (ONS, 2004).
It is important, when defining schools as serving HI, MI, or LI communities, to recognise that this label will not necessarily apply to every parent whose child attends the school. To further explore class variation between and within schools we collected data through the questionnaire survey to measure parents' National Statistics-Socio-Economic Classification. Specifically, the questionnaire used the ONS four-question self-coded method to gather data about the person defined by the respondent as the main wage earner (if any) in their household. Measuring the class of a household is always contentious, but here we privileged the class position of the main wage earner as they have the biggest influence on the economic circumstances of the household, and this allowed us to assess households consistently (rather than simply according to who completed the questionnaire). The subsequent threestage analysis and class collapse categorised parents into three groups: class 1-managerial and professional occupations; class 2-intermediate occupations; and class 3-routine and manual occupations (including never worked and long-term unemployed) (for details see ONS, 2012a; 2012b). The results are shown in table 1.
The survey was followed up with forty-five semistructured interviews which explored parents' attitudes in greater detail: fifteen parents were chosen from questionnaire respondents who volunteered in HI, MI, and LI schools to reflect both the mix of class backgrounds and types of family formation (dual/lone-parent households; working/not in paid work parents) evident in each group. The gender-neutral term 'parent' obscures some important trends in our data. In total 93% of those who returned the questionnaire were women, as were all of our interviewees (though on occasions they also chose to have another family member present for parts of the interview). In general, we use the term 'parent' in relation to the quantitative data to include the fathers who returned the questionnaire, but refer specifically to mothers where this is relevant in relation to the qualitative data. The interviews were all fully transcribed and subsequently coded in NVivo. To ensure anonymity, interviewees were allocated alpha-numeric identifiers for use in the storage, analysis, and publication of transcript data (eg, H1; M8; L12). These identifiers, along with parents own class position (C1, C2, C3), appear at the end of quotations from the transcripts in the text of the paper (NPW = no one in paid work). It is to an analysis of these data that the paper now turns.

Parenting education, social class, and local mothering cultures
Social class, place, and attitudes to parenting education Parenting education, like other aspects of parental support, has involved policies shaped by middle-class mores, where social scientists argue working-class parents have been exhorted to behave in more middle-class ways without reference to their wider material or cultural circumstances (Gillies, 2010;Klett-Davies, 2010). Our first aim is to explore what parents from different class backgrounds think about these policies, and how receptive they are to provision within their child's primary school. The results from the questionnaire survey (see table 2) show that the service is most popular with parents whose children attend LI schools. Here 82% of parents think that their primary school should provide parenting classes, whereas only 44% of parents with children in HI schools agree. This pattern also holds true when we look at parents' own social class (rather than school type), with those in class 1 being less likely to favour provision than those in class 2 or 3. Moreover, these statistically significant differences in attitudes to service provision are also reflected in parents' response to whether they themselves would use these services. Thus while social scientists have exposed the class biases inherent in the professionalisation of parenting (James, 2009), their concerns do not appear to be shared by working-class parents in this study who are more likely to favour parenting classes than their middle-class counterparts. It would seem that the idealised version of middle-class parenting on which the culture of parenting is based, is perhaps something to which working-class parents aspire (Perrier, 2010). Our second aim is to explore the importance of sociospatial context in shaping parents' attitudes to these services. Section (c) of table 2 demonstrates that parents whose children attend the same type of school tend to have similar attitudes to each other, regardless of their own social class. Thus within LI schools, for example, there is no statistically significant variation in parental attitudes by social class. What this starts to show is the importance of Notes: Data subject to | 2 analysis. Bold type indicates differences significant at 5% level. Normal type indicates differences not significant at 5% level.
place-based experience as the diverse people living in a neighbourhood, and sending their children to a particular type of school, develop similar attitudes to this service provision. Section (d) of the same table throws further light on this. Parents who share the same class position, it reveals, have significantly different attitudes to parenting classes depending upon the type of school their child attends. Professional/managerial parents, for example, are statistically more likely to favour parenting classes if their child attends an LI than an HI school. Thus, while social scientists are right to argue that parenting cannot be abstracted from its social context because class does indeed matter, these quantitative data also show that class cannot be abstracted from its sociospatial context because the neighbourhood in which you live, and the school your child attends, matter in the development of classed attitudes to parenting education.
In the next section we go on to use the qualitative data to explore the reasons behind class differences in parents' attitudes to parenting education, and draw out the importance of place in the ways these attitudes emerge in different sociospatial contexts.

Higher-income schools
We begin our exploration of the qualitative data by considering the sociospatial context in which parenting is taking place. Mothers whose children attended HI schools were generally very positive about the areas in which they lived. They considered their children to be extremely lucky to be growing up in a materially privileged environment with few social problems, and thought they faced relatively few challenges growing up: " I don't think they have any! I think they're extremely fortunate to be able to live in this kind of environment, have all these opportunities, be run around all over the place to do whatever they want to do … they are extremely fortunate" (H12:C2). " I don't think they have many challenges really, just the natural learning in life that everything, so you're not always equal, and the frustrations of life of just being a kid that everybody goes through really" (H4:C1). When problems were identified, these were the flip side of living in this privileged environment (Katz, 2008). Some mothers were concerned about materialism and the spoiling of children, and a small minority were concerned about the competitive nature of the area which placed children under pressure to excel (Vincent and Ball, 2007).
The general shared satisfaction with the neighbourhoods in which they were raising their children fed into a relatively low demand for parenting classes. Mothers thought children in their area were generally well brought up, and stated that their own were not posing any significant problems, and thus that there was no need for parenting classes: " I'm not sure that many people have that many serious problems with behaviour in the school, I think generally behaviour is of a pretty good standard within the school, so I wouldn't imagine there'd be a big need for that" (H6:C1). " Me personally, I don't think we have any major problems with the kids, I mean they are pretty grounded. … So I wouldn't be interested" (H3:C1).
They were also confident that if problems did arise they would be able to use their own local social networks to find answers. Previous research shows that middle-class mothers are more likely to turn to their friends for support than are their working-class counterparts (Edwards and Gillies, 2004). These friendship networks can be online (Madge and O'Connor, 2006), but are often anchored in local neighbourhoods which develop distinctive parenting cultures (Foy-Phillips and Lloyd-Evans, 2011;Holloway, 1998). In this case, these networks act as a resource in supporting parents, usually mothers, in the raising of their children. This support, which feeds into a rejection of parenting education, is facilitated through face-to-face contact at neighbourhood parenting locations: for example, when dropping and collecting children from school (Witten et al, 2007). It rests, as the mother below explains, on an assumption of shared family values that cross-cut other differences between women (eg, working/not in paid employment): " I wouldn't sign up to it [parenting classes] … because I don't think that I struggle particularly, and when I do struggle, because we all struggle … . I would use my friends as a network to resolve those issues and talk that through, who know me … . I think that most people have got a network of friends and would do it in that way … other mums are very similar to me, and whether they work or don't work, they've got a network of friends and they use that. And because of the type of values we've got, the family and the kids are important, so it is a topic of conversation" (H15:C1) Moreover, in addition to this availability of day-to-day support from a locally networked community of mothers, parenting education was not seen to be required because parents were envisaged as competent in negotiating access to help for their children if specific needs arose (Edwards, 2010): " I think the parents of the children that go to [our school] are quite able to articulate the issues with teachers or other professionals; I think they're the sort, a group of parents who would go and seek the help and know where to find it" (H2:C1). Notwithstanding parents' understanding that their children face few challenges, and that mothering networks were there to provide support, the quantitative data did show that 35% of parents with children in HI schools would go to parenting classes. This was presented by these parents as a sign of self-confidence rather than concern: they would try parenting classes as they were assertive and would try most things.
" I'm fairly sort of open and bold about things like that" (H12:C2). Moreover, in a practice akin to middle-class mothers' accrual of bridging social capital through participation in parent teacher associations (Bagnall et al, 2003), a minority of mothers who lacked mothering networks (for example, because they had moved since having children) saw parenting classes as a strategic way to forge links with others and discuss their views on parenting.
The picture to emerge in HI schools is thus one where mothers facing relatively few challenges, and supported by networks of other mothers, express low demand for parenting classes. Their largely middle-class experiences and attitudes are constituted in and through neighbourhood space, spaces which themselves are also shaped by wider class relations. Mothers' judgments that children face few problems are influenced by their positive local environments; the networked mothering cultures which provide support depend on a local sense of sameness and are rooted in neighbourhood space. Experience in place is thus important in shaping classed attitudes to parenting class provision. Indeed, the importance of place-based experiences in the sociospatial construction of classed attitudes explains why those from other social classes whose children also attend these schools share the same attitudes as their more middle-class counterparts here.

Middle-income schools
Mothers whose children attended MI schools tended, regardless of their own class background, to identify more challenges in the raising of their children than those in HI areas. The problem to which they paid greatest attention was interaction with, and the influence of, 'other people' (although finance was also a cause of concern). Table 1 shows that in class terms MI schools have a mixed intake, with 41% in professional and managerial employment, 15% in intermediate occupations, and 44% in routine or manual occupations (including never worked and long-term unemployed). These parents did not assume that everyone in their area had a similar outlook to themselves. On the contrary, the variability within their neighbourhood, and between their neighbourhood and contiguous communities, was a cause of considerable disquiet. Some mothers were clear that they were not keen to mix with all the different types of parents in their neighbourhood: " I don't want to get myself involved with some people, I just, those I need to know, I need to know, and those that I don't, just stay out of the way" (M12:C1). " There's a few that I will give a wide berth to on the street because I don't want to get involved. There's a few on the street that are lovely" (M9: C2). Equally they were certain that they did not want their children to mix with all the other local children. This was in part to protect their 'Apollonian' children from the risk of other children's bad behaviour, but more importantly was designed to ensure their children would not be led astray by 'Dionysian' others (Skevik, 2003): " I'm terrified he'll get in with the wrong crowd, I mean I see some of the kids around here and I think mmm I really don't want, oh God that sounds so snobby doesn't it, I really don't want him socialising with some of them" (M13:C2). " [Daughter], she goes on park doesn't she, but I suppose you have to let them, I'm not too happy about it sometimes, because if you saw some of the people on [named] Park you'd understand, but…!" (M10:C3). The majority of these mothers using MI schools did not themselves want to use parenting classes. Many do not think they need this service as they feel confident in their parenting abilities and are not experiencing significant problems with their children. To this extent their position is similar to mothers using HI schools, but it differs in that, while those mothers conformed to a middle-class pattern of extolling the virtues of friendship support networks (Edwards and Gillies, 2004), this more mixed group did not, sometimes arguing instead that they had family resources they could draw on should any problems emerge: " I've got my family around me, you know I'm not, I like to think that I'm quite a good mum" (M7:C1). Moreover, the qualitative data revealed a greater willingness amongst mothers here to access parenting classes if they started to experience problems, a factor which in part might explain the greater numbers of parents here who think parenting classes should be provided (see table 2). This openness to parenting education was particularly evident amongst those, such as the mother below, with younger children: " Yeah, if my child was going through something, you know, any advice is welcome isn't it? … Yeah if I was stuck and didn't know what to do or, you know then yeah I'd be open to trying new things" (M15:C2). In this context, there was a considerable level of confidence displayed about the use of parenting classes. Those mothers who had good experiences of using them in the past, and those who thought they might use them if they had problems, displayed a self-assured attitude to accessing the service. Concerns have been raised elsewhere that parents' fear of the unknown, and fear of being judged by others, can limit uptake of parenting education (Apps et al, 2007). This group, however, appear largely confident in meeting other parents in institutionalised spaces, and represent their desire and ability to do so as a mark of good mothering, a mark which they use to set themselves apart from 'other' mothers: " I've done most of the things [parenting classes] they've offered … people that tend to put themselves forward to do these things, are the people that are already bending over backwards to make sure the kids have got everything" (M3:C3). " The thing about parenting classes and parenting courses is that I've found that the people that actually attend parenting courses are the ones that actually don't need parenting courses! … some of them would definitely benefit from more information I think, or more education on parenting, but as I say I don't think those are the sort of people that would attend courses, they'd see it as some kind of stigma I think" (M5:C1). This process of 'othering' those who do not attend parenting classes enables mothers to maintain a moral sense of self through contrast with real or imagined others, reinforcing in their own minds that they are good parents, even in cases when they themselves have accessed external advice. As Perrier (2010, page 25) argues, "it seems that peers or imagined other mothers are crucial for the persistent circulation of good mothering discourses." Moreover, these discourses which identify 'other' parents as being in need of parenting advice are a second factor which in part explains the greater number of parents who think parenting classes should be provided in this area than would themselves intend to use them (table 2).
In the MI school context we do not see the emergence of a sense of commonality amongst mothers who share similar class positions, or indeed of a coherent grouping who might reasonably be labelled as 'lower-middle-class' or ' upper-working-class' mothers. Class continues to matter in attitudes to parenting, but it is experienced here through local mothering cultures which are constituted through the heterogeneity of class backgrounds in these areas, rather than through the dominance of one particular class grouping. The consequence of this local social mixing is that mothering networks play a much less important role in this area, but the provision of parenting classes is more popular, with participation often being articulated as a mark of good mothering compared with stigmatised local 'others'.

Low-income schools
The sense of variability within an area that emerged from parents who use MI schools was also apparent amongst those from all social classes who use LI schools. Mothers were clear both that some parts of their neighbourhood are 'better' than others, and that within locally demonised streets there are both 'good' and 'bad' families: " [T]hese on the outskirts is quite [alright], you know here, but go into the middle of that estate it's really not good … a lot of drinking, drugs, arguments, fighting, there's litter, it's not nice!" (L1:C1). " I mean not everyone's like, I mean there is people that obviously, you [partner] were saying like oh you can spot a smack rat now, but not everyone is, maybe they're in that sort of situation you know, they live on like [named street] and it's not very nice, but they are like us in the sense that they want to get out but they can't sort of thing" (L10:C2). In this context, one of the major concerns parents articulated was fear about children being led astray by others who were seen to behave badly in the neighbourhood (Vincent et al, 2010): " [N]ot to be led astray is a big challenge, especially for my eleven year old, not to be led astray by other children doing what they shouldn't be doing really … . Some of the kids smash the windows in the bus shelter; it's forever being smashed, and spray painting bins and setting bins on fire" (L9:C3). This was considered particularly problematic as many parents thought there was little for children to do, both because of a lack of local facilities such as well-maintained parks, and because in this generally low-income area they struggled to find money to pay for leisure activities (Power, 2010).
Notwithstanding the challenges in their locality, some parents as in HI or MI schools felt confident they were doing well in raising their children: " I don't think we struggle, they're pretty good kids" (L5:C1). " I think I'm doing a good job with them really!" (L15:C3-NPW).
However, this was not reinforced by a sense that everyone else in their area shared their values and was equally successful. Rather, mothers thought that many 'others' in their areas did not care appropriately for their children: " To be honest, I think half the parents round here just tend to let their children go out and think well bugger off out, just to get them out the way really. They don't really care what they do" (L9:C3). " There's some parents, they just basically have kids for money [state benefits] and just can't be arsed to look after their kids … they can't be bothered to do anything with them, and not bothered whether they run round streets or not" (L2:C3-NPW). While this assessment of 'other' people's poor parenting clearly makes allusions to wider discourses about the deserving and undeserving poor in the context of benefit entitlements, it is noteworthy that the stigmatisation of lone parents was not a defining feature of this local discourse, nor in these primarily white areas was the discourse explicitly racialised.
This negative judgment about 'other' people's parenting is, however, one reason why support for the provision of parenting classes tops 80% amongst parents using LI schools, a demand for parenting support that is also seen in other LI areas (Power, 2010). This concern is a real one, and living in an area with significant antisocial behaviour certainly does lead to support for parenting classes as 'others' are seen to require it. However, the process is not simply an objective one in which 'good' parents identify 'poor' parenting in others. Rather the discourse has a very wide social currency and it is also used by those to whom others might seek to apply it. Regardless, this belief that there needs to be more parenting education would appear to be shared by these schools: this is evident both in the attitudes of Hortonshire headteachers (Holloway and Pimlott-Wilson, 2012), and the national picture where parenting support is more likely to be offered in schools with relatively high levels of deprivation (Wallace et al, 2009).
This demand for parenting classes is coupled with a fairly strong assertion that those 'others' would not, or indeed do not, go to parenting classes offered as they, unlike the state, think parenting is a private matter or would fear being judged. There is indeed some resistance to parenting classes in the area, especially from those with older children. As this mother explains, she would not want parenting classes as she sees herself as competent and would resent interference: " [Parenting classes] would just get my back up, they'd tell me how to bring my kids up you know, I'm on my third here, I don't need anyone telling me, I've got through the other two alright, mind your own beeswax [business]!" (L8:C3-NPW). It is important to note, however, that some of the resistance to parenting classes in this area is because a range of parents have been invited or felt compelled to attend classes when their children have experienced health and/or social problems (Barrett, 2008). This type of targeted parenting class, rather than universally accessible ones provided through school, to some extent stigmatises all parenting classes as they become associated with compulsory state involvement and parental failure.
Nevertheless, over half of the parents using LI schools reported through the quantitative survey that they would use parenting classes. These mothers explained that parenting classes could be useful if you were experiencing problems with your children, or started to do so in the future. Once again, this was presented as a sign of good parenting as mothers were willing to do this for their children: " I mean I'd probably only ever do sommat [something] like that if I were having a major problem with one of the kids and probably needed help sorting it out … . I think if you care about your kids then you'll do the most you can, and there's probably some people down at school that probably don't care as much about their children" (L13:C3-NPW).
However, despite high suggested figures for usage in the questionnaire, our work with providers shows that it is not always that easy to fill parenting classes. Our results suggest that this is both because of the stigma attached to these classes, and because some mothers lack self-confidence, meaning they feared attendance (Apps et al, 2007) with some remaining isolated from sources of support (Harknett and Hartnett, 2011). In these areas, then, we see how a class culture is made in, and experienced through, place. The majority of parents using LI schools were either in routine/manual occupations or not in paid employment. Their experiences of living in what were described by interviewees as 'deprived areas' (L7:C3-NPW) where antisocial behaviour is a significant problem led to significant support for parenting classes as other people were seen to need to learn to parent better. To cast these mothers simply as cultural dupes who have bought into neoliberal discourses about individual responsibility for social outcomes (Raco, 2009), placing the blame for neighbourhood problems solely at the feet of other people's poor parenting, is inappropriate, however, as they were also cognisant of the impacts of widespread poverty, unemployment, and lack of services on their areas. Nevertheless, while academics have rightly criticised the professionalisation of parenting for imposing middle-class mores on working-class parents, amongst mothers in LI schools this middle-class model is embraced, not least because it can be strategically deployed to assert individual maternal respectability through distance from 'other' mothers in a climate where working-class parenting is commonly devalued.

Discussion and conclusion
The purpose of this paper has been to direct increased geographical attention at new and extended forms of teaching and learning which are developing under contemporary neoliberal educational reform. Our particular focus has been on parenting education (Shulruff et al, 2009) through which the state is not simply seeking to educate individual citizen-workers but rather, given the importance placed on parenting in contemporary policy discourses (DfES, 2007a;DfE and DoH, 2011), to shape the contexts in which future citizen-workers are raised. The recent professionalisation of parenting has been critiqued for imposing middle-class mores on working-class people, whilst ignoring differences in the social context in which they parent (Gillies, 2010;James, 2009;Klett-Davies, 2010). Our aim has therefore been to explore how parents of different social classes feel about parenting education and, given the limited attention paid to spatiality in research to date, how these attitudes are influenced by the sociospatial context in which they parent.
Our empirical findings reinforce social science research which stresses that parenting is a class issue, as middle-class and working-class parents' strategies are shaped by and for different social contexts (Edwards, 2010: Gillies, 2010James, 2009). This is evident in our quantitative data which show that attitudes to parenting education vary significantly with social class, with middle-class parents being statistically less keen on its provision and use than their working-class counterparts. Crucially, our analysis also takes understanding one step further by highlighting the links between classed attitudes and the sociospatial context in which people are parenting. Specifically, our data also demonstrate that place matters because parents of the same social class express significantly different attitudes to parenting education depending on the nature of school their child attends. Parenting is a class issue, but place also matters in the (re)production of classed attitudes to parenting education.
The qualitative data shed more light on the ways classed attitudes to parenting education emerge in particular places (places which are themselves shaped by wider class relations). Local class-based cultures of mothering influence individual mothers' attitudes to parenting education. Amongst parents using HI schools, for example, networked mothering cultures, which are underpinned by a sense of neighbourhood satisfaction and sameness, act as a source of support and reduce demand for parenting classes. Mothering culture in the far more heterogeneous MI schools was characterised by more individual/family based approaches to mothering; some mothers were self-confident in access parenting classes for themselves, but we also start to see a belief in the need for this provision for 'others'. Mothers using LI schools are generally less connected to others, and individual mothers are apparently more open to parenting classes both for themselves, and importantly for local 'others', in part as parental education is seen as a potential remedy to local social problems.
These local mothering cultures are class based in the sense they reflect the opportunities and challenges of the different class groupings that constitute their areas; however, placebased experiences are crucial in the reproduction of what are sociospatial relations. For example, the networked mothering culture in HI areas depends upon a sense of neighbourhood satisfaction and sameness; the everyday experience of antisocial behaviour reduces trust in LI areas and underpins support for parenting classes for local 'others' who are seen to need it. Indeed, it is the importance of these local experiences which means parents who are not from the dominant class grouping in their neighbourhood come to share the values and opinions of others parents around them. It is this which explains the enthusiasm for parenting education amongst middle-class mothers using LI schools, and conversely the rejection of parenting education amongst working-class mothers using HI schools. Thus, local class-based cultures of mothering can be defined as class cultures, even as we recognise that they are not neatly confined to members of a specific class grouping, with individual experience being shaped by their neighbourhood context, places which we have shown matter in the reproduction of these social-spatial relations.
Places also matters at another level precisely because localities are characterised by different local class-based mothering cultures (Foy-Phillips and Lloyd-Evans, 2011;Holloway, 1998;Witten et al, 2009). These cultures not only shape individual experience but also the nature of neoliberal policy implementation (cf Braun et al, 2010). The networked mothering culture in HI areas reduces demand for parenting education and the viability of provision, even as some mothers feel they would benefit from this. Equally, processes of establishing maternal moral respectability by denigrating 'other' mothers (Perrier, 2010;Vincent et al, 2010) can lead to the stigmatisation of parenting education in LI schools, reducing social accessibility in a context where more than 50% of parents want to use the service. The professionalisation of parenting, in this instance in the guise of parenting education being offered through primary schools, is being differentially embraced, tolerated, and/or resisted through the local mothering cultures in different socioeconomic communities.
This study of attitudes to the provision of parenting classes through primary schools, though interesting in itself for the insights it provides into the nature of social class and local mothering cultures, also has wider implications for how we think about the restructuring of education in neoliberal states (Butler and Hamnett, 2010;Hanson Thiem, 2009). Firstly, it emphasises the importance of considering new educational functions (Jupp, 2011;Wainwright and Marendet, forthcoming) which are emerging as governments seek to shape future citizenworkers in new ways. A characteristic of neoliberalism, particularly in its roll-out formations seen in this paper in England, has been a move away from envisioning the family as a private unit, with this instead being seen increasingly as an acceptable focus for policy intervention (Edwards, 2010). The consequence of these contingent forms of neoliberalisation is that the boundaries between family life and state responsibility in the arena of social reproduction are shifting (Ansell, 2008), and geographers need to trace the implications of this for parenting and educational provision in different spaces and places.
Secondly, it demonstrates the importance of taking the subjects of education policy seriously (Holloway et al, 2010). The importance of education on the political agenda (Jenson and Saint-Martin, 2006) means policy appears to be in a permanent state of upheaval.
Like other areas of neoliberal policy reform (MacLeavy, 2008), this paper shows that it is crucial that we explore how the subjects of that policy, in this case parents, feel about these developments. This is not simply because those policies can have important outcomes for individuals and groups, but also because the subjects of policy can make a difference to the success or otherwise of a state project. Neoliberal policies are not simply implemented : as this paper has shown, the responses of those who are its subjectsincluding enthusiasm, tolerance, and resistance-matter to how education emerges in practice in different sociospatial contexts.
Finally, the paper highlights the importance of examining the wider geographies of education, rather than focusing narrowly on the different learning contexts in which it is delivered (Brooks et al, 2012). In this case the paper demonstrates that the community contexts in which parents operate matter to their attitudes to a specific form of educational provision. The significance of social class has been a longstanding concern in both geographies of education, and broader educational research. This paper shows that these links between social class, the wider sociospatial context, and educational aspirations, experiences, and achievement in a range of forms of provision are worthy of greater attention in future research into geographies of education. Moreover, in highlighting the importance of place in mediating the nature of social class the paper demonstrates the need for educational research beyond the discipline to attend to the importance of geography.