Class, capital, and school culture: Parental involvement in public schools with specialised music programmes

Abstract This study examines parental involvement in urban public schools, focusing on how parents in organised school-centred networks support, navigate, and negotiate from their different social positions. The multiple case study of specialised music programmes provides insights into parent strategies and behaviours in intermediate practices between school-based socialisation and extracurricular activities largely run by the parents associations. The paper draws on data from in-depth focus group interviews with members of parents associations in socially, culturally, and historically different schools. Findings demonstrate that parent approaches to specialised education and their modes of involvement vary according to social class, resources, and school culture. There are class-based differences in parent strategies and the way their collective symbolic capital is used in policy negotiation. However, relationships between a parents association and the school administration are generally regulated by the social and cultural history of a particular school.


Introduction
In neoliberal education systems regulated by market principles, parents are expected to use their freedom of school choice to customise their children's education. Critics have argued that the educational marketplace entails inequality since middle-class parents are more likely to take advantage of their privileged resources in terms of cultural capital and social benefits. Thus, studies frequently conclude that policies involving educational choice contribute to the reproduction of social stratification (Ball 1993(Ball , 2003Vincent, Braun, and Ball 2010). Similar conclusions have been drawn by research on enrolment in extracurricular activities (Bennet, Lutz, and Jayaram 2012;Lareau 2011;Vincent and Ball 2007). Such organised out-of-school activities largely attract middle-class parenting styles, where parental involvement both inside and outside school is considered morally desirable.
A considerable literature in the sociology of education has explored the relationship between engagement by parents in their children's education, extracurricular activities, and the child's academic achievement (e.g. Covay and Carbonaro 2010;Dumais 2006;Kaufman and Gabler 2004). While such studies have concentrated on the individual level in parent-child relationships and emphasised the impact of extracurricular activities on academic performance, research has rarely considered patterns of collective involvement in local parents associations in determining school-based extracurricular activities, especially with regard to power relations between those organisations on the one hand, and school administrators and education policy makers, on the other. However, some exceptional studies may be mentioned (see literature review below and Cucchiara and Horvat 2009;Horvat, Weininger, and Lareau 2003;Lareau and Lopes Muñoz 2012). These authors have shown the importance of interrelating research concerns about parental involvement and social networks with issues of parent preferences and educational strategies.
The present paper examines parental involvement in urban public schools, focusing on how parents in organised school networks support, navigate, and negotiate from their different social positions. We focus on the phenomenon of specialised audition-based music programmes in Swedish urban public schools, since enrolment in such study tracks appears to be related to both social capital mechanisms and codes of cultural capital (Gaztambide-Fernández and Parekh 2017;Gaztambide-Fernández, Saifer, and Desai 2013;Lilliedahl 2020). Moreover, this type of specialisation implies social interactions in intermediate practices between school-based socialisation and associated extracurricular activities run largely by the parents association. These 'music classes' are usually organised in parallel to a school's regular classes and are found in vastly different social environments, ranging from the most prestigious and highly selective specialist schools to low-performing ones in disadvantageous areas.
In the present article, we view parent strategies and involvement in parents associations together with how parents use their collective symbolic resources to support their children's education. Our aim is to explore why families in different social and cultural environments choose to apply to specialised music programmes, how they operationally and strategically support or negotiate with local school management in a variety of ways, and whether their patterns of preferences and actions interrelate with different types of school cultures.
We believe our findings provide a new understanding of how differences in parent behaviours in school-based networks are conditioned by social class, symbolic resources, and school culture. We find significant relationships between parental involvement in extracurricular activities and the preferences and expectations on which they base their educational choices.

Literature review
Since educational choice is dependent on cultural capital, such as family resources, research has shown that middle-class parents are in a privileged position to impact their children's education (Ball 1993(Ball , 2003Goldring and Hausman 1999;Lareau 2000;Vincent, Braun, and Ball 2010). Those parents are usually eager to take their children's education in hand, while the child's view is more likely to be decisive in working-class families (Reay and Ball 1998). Researchers have thus argued that the education marketplace entails inequality, since advantaged parents are more likely to make use of their privileged resources in order to transfer cultural capital and reproduce social advantage.
Similar conclusions have been reported by studies of organised extracurricular activities. Researchers assert enrolment in such activities is associated with middle-class lifestyles (Covay and Carbonaro 2010;Lareau 2011;Vincent and Ball 2007). While middle-class parents typically see extracurricular activities as a means of 'concerted cultivation' , working-class and poor parents are more likely to think in terms of notions of 'natural growth' , typified by unstructured after-school activity (Lareau 2011). Although extracurricular activities often require enrolment fees, participation has been associated with cultural preferences more than economic circumstances (Weininger, Lareau, and Conley 2015).
Social dispositions and cultural capital have also been found crucial in making educational choices, including the decision to apply to schools with high status (Lareau, Evans, and Yee 2016;Weininger 2014). It appears to be of importance whether parents are capable of distinguishing between schools. Moreover, parents have to be familiar with the 'rules of the game' , including filling out applications, observing submission deadlines, and taking admission tests. Choosing an arts programme is thus likely to be related to one's cultural capital and social disposition, which suggests that such programmes may be self-perpetuating by continually attracting a homogenous group of resourceful families (Gaztambide-Fernández and Parekh 2017).
In the context of educational choice, pursuing specialised arts tracks may be considered morally justified since such programmes offer extracurricular opportunities viewed as organised cultivation through artistic activities (Gaztambide-Fernández, Nicholls, and Arráiz-Matute 2016;Saifer and Gaztambide-Fernández 2017). In this sense, concerns with specialisation and arts education may be understood as the efforts of parents to invest in and remain emotionally close to their children (Reay 2004;Stefansen, Smette, and Strandbu 2018). However, there may be other reasons for applying to an arts class, such as to ensure a safe environment -something of great concern to anyone who has been in a vulnerable school situation (Bennett, Lutz, and Jayaram 2012;Gaztambide-Fernández and Rivière 2019;Vincent, Braun, and Ball 2010).
Class-based differences in parenting style are also reflected in parental involvement. Parents differ in the number and types of resources they control, and how they are able to support their children (Lareau 2000(Lareau , 2011. Middle-class parents generally have a greater capacity for interacting with their child's school and having their way. They stay informed and are more likely to visit the school to negotiate in matters concerning their child's education . Class-based differences impact home-school relationships. While working-class parents commonly consider their child's school as an autonomous institution, completely separate from the home, middle-class parents are likely to be more confident in interacting with teachers and school personnel (Lareau 2011). Middle-class parents are typically better organised to influence their children's school experiences (Lareau and Lopes Muñoz 2012). In 2003, Horvat, Weininger, and Lareau found that by contrast to the kinship ties that bind working-class families, middle-class parents tend to build homogenous networks that arise from their children's organised activities. Such networking enhance their ability to intervene in school matters. As a consequence, middle-class parents are more likely to secure the aid of professionals and collectively dispute the school's authority, while their counterparts are typically more timid, family-centred, and less likely to question the school administration.
Organised parent-school relationships can lead to clashes over issues of authority, control, and priorities. There may be interaction between groups of parents and school administrators on a structural level, indicating that the individuals are embedded in an institutional framework (Lareau and Lopes Muñoz 2012).
Despite numerous studies demonstrating how social-class differences determine parental involvement in education, there is research that challenges the assumption that parents of higher socioeconomic status are more likely to be involved in their children's education (e.g. Tan 2019). Results from North America and England, where working-class parents are less likely to intervene in their children's school choices, may contrast with those from the Nordic countries (Hegna and Smette 2017). In order to take cultural differences into account, the relationship between social class and parental involvement needs to be viewed in different cultural contexts.
In proposing that class and culture should not be considered separate categories, Erdreich and Golden (2017, 61) suggest that they be seen as 'social processes' for a better 'understanding of how parental involvement is shaped' . Since the social and cultural history of a particular school may influence how parents are differently engaged (Cucchiara and Horvat 2009;Lareau and Lopez Muñoz 2012), parental behaviour may also be conditioned by a school's culture Bernstein 2000;Power et al. 2003). Specialised arts programmes themselves vary considerably from school to school depending on the social context, just as different schools attribute different meanings to the concept of specialisation (Gaztambide-Fernández, VanderDussen, and Cairns 2014). Thus, parental networks may serve a variety of purposes, especially in disadvantaged school neighbourhoods (Li and Fischer 2017). Moreover, there may be variations within the middle class that should be taken into account for a nuanced description of how relationships between class, parental involvement, and culture are interconnected (Bernstein 1975;Power et al. 2003;Power and Whitty 2002;Reay, Crozier, and James 2011;Vincent and Ball 2006). Consequently, further study is needed on parents' educational strategies and parental involvement that takes the particular school context into account relative to differences between and fractions within social classes.

Theoretical framework
In bringing the notions of class, culture, and parental involvement together, we have sought to develop an integrated framework based on Bourdieu-informed theories and an understanding, following Bernstein, of how the field of education is regulated by principles of classification and framing.
What Bourdieu (1984) defines as habitus refers to a parent's disposition due to personal experience and past actions, that is, the patterns of preferences and behaviour they exhibit on the basis of their social and cultural background. Habitus thus determines one's parenting style and strategies for the 'structuring of school experiences' (Bourdieu and Wacquant 1992, 134). Habitus is of particular importance since it refers to differences in the material and symbolic resources -capital -parents use in their relations to the field of education (Bourdieu 1984;Bourdieu and Wacquant 1992;Lareau 2011).
Cultural capital refers to those forms of knowledge and education that parents possess, typically their familiarity with the fine arts and the socially distinguishing principles of cultivation (Bourdieu and Passeron 1990). Our interest here is in how cultural capital is 'embodied' in parents' educational strategies, including their 'informational capital' , which gives them the advantage of being informed about the educational system (Bourdieu 1986;Bourdieu and Wacquant 1992).
Social capital works through parental actions in relation to the field of education (Bourdieu 1984). The context-specific value of social capital suggests a research focus on the 'rules of the game' , along with how parents navigate to improve their children's school experiences (Lareau, Evans, and Yee 2016). Our interest here is in parents' qualitatively different ways of networking in building and reproducing parents associations and interacting with the school administration. In order to do so, parents may use their professional or bureaucratic assets to support their efforts (Savage et al. 1992).
By analogy with social capital, cultural and economic capital should be analysed as resources activated within a specific field. In respect to the latter, we have focused on how economic resources are utilised as symbolic capital, the latter being any type of capital (cultural, social, or economic) that becomes recognised as symbolically valuable in a given context. In sum, the power of a parents association resides in its economic resources, its social networks and relationships, and the dispositional relations of its members' proclivities. These forms of capital mediate parental influence and involvement.
The notion of field may be used to conceptualise the educational market and the relationships between parents' dispositions, their parenting behaviours, and educational institutions (Bourdieu 1984). However, as Swartz (1997) has pointed out, Bourdieu's concept of field was not originally intended to demonstrate how interactions are regulated in an institutional setting, but rather how specific social worlds are formed and reproduced on the basis of habitus and capital. We prefer Bernstein's way of conceptualising the field for two reasons. First, the issues we address are less about how individuals and elites struggle to capture symbolic capital and thereby achieve a desirable position within a culturally-bound social field; rather, they involve how middle-class factions are regulated by (and in turn regulate) the 'rules of the game' in the field of public education (Bernstein 1975). Second, since different school cultures may be 'open' or 'closed' to varying degrees, these structural differences must be taken into account to understand why parents associations behave in different ways (Power et al. 2003). Bernstein's (1975Bernstein's ( , 1990Bernstein's ( , 2000 conceptual toolkit, principally based on the strength of social relations in education (classification and framing), can complement Bourdieu's framework in allowing one to systematically analyse interrelations between social class, parental involvement, and school culture Power et al. 2003). The Bernsteinian concept of a school's expressive order reminds one of what Reay et al. define as institutional habitus, that is, the impact of a school's disposition and organisation as determined by its historical, social, and cultural background (Reay, Crozier, and Clayton 2010;Reay, David, and Ball 2001). Thus, it becomes equally important to describe class-based differences between schools as differences between parents, making the term middle-class schools more telling than middle-class parents.

The study
The Swedish comprehensive education system includes urban public schools with audition-based specialised music programmes where students are given extensive musical instruction, including extracurricular activities largely arranged by the local parents association. The specialisation implies an adjusted curriculum for which parents and their children may apply in conjunction with choosing a school. Most often, such programmes are organised in parallel to a school's regular classes. In this way music classes comprise a subcategory within the quasi-market of public and private schools in Sweden.
Those schools offering specialised music education are usually found in large municipal school systems and in medium sized towns. Some are located in affluent neighbourhoods and so are likely to attract upper-middle class families, while others are situated in poorer or disadvantaged areas and enrol a social mix of children from mostly lower-middle class or working-class families and ethnic minorities. Taken together, these schools reflect a spectrum of student achievement levels and parental educational backgrounds ranging from high to low.

Method
Our research employed a multiple case study methodology based on semi-structured focus group interviews with members of the governing board of the parents association at seven school districts. Since we were interested in the collective features of the relationships between parents associations and the codes of particular schools, we categorised associations broadly based on a school's average cultural-social class composition. The schools were selected according to official Swedish statistics on educational results and parental education level (SIRIS). In the schools we designate as upper-middle-class, the average student achievement level in grade 9 was among the top 5% of all schools in Sweden, and the parents of those students were among the top 5% best-educated in Sweden: 80% to 100% held a higher education diploma (B.A. and above). The lower-class schools in our study included some of the lowest-performing schools in Sweden (bottom 7.5%), with a majority of the parents lacking post-secondary education. The average middle-class schools encompass the broad spectrum in-between, with both relatively good and relatively poor student outcomes, including variation in parental education levels. Thus, our cases included wealthy as well as poor neighbourhoods, academically strong and weak areas, and high and low performing schools, a fair representation of the geographical distribution of specialised programmes in Sweden (a breakdown given below). However, due to Swedish research guidelines and regulations, data on school results and parental educational levels cannot be given in detail.
The school called A is an upper-middle-class school located in an upscale, densely populated area in one of the country's major cities. It offers a number of subject specialisations, including a long-established music track supported by an active parents association.
School B is the other socially-privileged school in this study. It is centrally located in a city that houses a major university. Unlike A, however, school B's approach to specialisation in music is very relaxed, and virtually unknown in the surrounding community. The parents association is loosely organised.
Schools C and D are located in the same metropolitan area, but in socially diverse neighbourhoods. Their academic outcomes vary considerably: whereas C is a relatively high-achieving school, D is one of the country's lowest. The latter has an overrepresentation of students from immigrant backgrounds and is in a neighbourhood the police has classified as vulnerable. In both cases, the programmes and the parents associations are relatively newly-established.
Located in a typical working-class neighbourhood of a medium-sized town, school E is locally known to be low-performing, but in national comparisons it is in the lower middle tier. Both its music programme and its parents association have a comparatively short history.
In the case named F, the well-established music programme and parents association comprise two schools located in different districts of the same industrial city. Both are in an area with a relatively high population density. One a typical middle-class school, the other, although similar, is in a neighbourhood that until recently was classified as vulnerable.
School G has several middle-class characteristics, such as its average student academic performance, parental education level, school location, and general demographic data. Its programme and parents association have a comparatively long history.
Participants were solicited by emailing the chair of the parents association. Several associations expressed their interest. The seven cases we selected of the 17 parents associations we approached were willing to participate and meet for an interview.
One interview with each focus group was conducted in 2019. The interviews were held in whatever locations suited the participants. In most cases, the association was represented by some members of its board. Three to 10 parents took part in each interview, which lasted from 45 to 80 minutes (average time 70 to 75 minutes). The interviews were characterised by open-ended conversations focusing on parental expectations and strategies on the one hand, and on practical matters on the other. Based on digital recordings of the sessions, transcriptions were made and translated into English. All names used are fictitious.
The subsequent analysis was an abductive act, with interactions between the categorisation of data and the theoretical framework. Through repeated examination, a variety of themes emerged. These themes were then condensed into different parental preferences and expectations, and variations in the architecture and functionality of the associations.

Findings
We found strong evidence that parental approaches to specialised programmes and modes of involvement varied according to social class, cultural preferences, and the historical context of the school.

Differences in parental preferences and expectations
Parental strategies appeared to be of crucial importance in their children's seeking admission to upper-middle-class schools, whereas in lower-middle and working-class neighbourhoods, interest in applying to music programmes was usually initiated by the children themselves. However, some middle-class parents living in disadvantaged areas said that the possibility of joining a well-functioning specialised programme offered their children a way to distance themselves from regular classes where disciplinary problems were a constant threat to learning.
At school A, most parents stated that they had played a prominent role in their children's school choice. Their decision to apply for the specialised programme was based on the school's reputation and statistics on student academic performance. Moreover, the parents' own educational backgrounds dictated certain preferences. This became especially clear when they were asked to put into words why they had chosen the school in question.

Transcript 1 (school A)
Alice: I was hoping for improved academic achievement and a more substantial education [for my child]. Bea: I think that culture is important. Singing and dancing has given me a lot. I believe it enriches the children, that they need that stimulus. It's good for everything. Then again, I am one of those parents who investigate what grades the kids get. At A it was [good]; everything spoke for the school. Cesar: For me it was crucial that my wife had studied at [names a similar specialist school]. That was where the idea came from, but I wasn't uninterested in or ignorant of music, so we thought it would be a good idea [to apply]. I have also looked at school performance, and I thought it was good [at A]. David: My wife had studied at the [mentions the same school as Cesar], too, and we are generally interested in music.
[…] Music is important since […] it acts as a kind of cohesion, a focus. Eric: Both my wife and I have studied at a school with specialisation in music. […] We are a singing family, with a great interest in music. We obviously thought our daughter was going to apply. She really wanted to. Felicia: I had a 'songbird' at home who generally liked music, which both sides of the family does also. Still, it was on my initiative that she applied [The mother continues to describe how she gradually persuaded her daughter to apply, take the admission tests, and finally accept the acceptance offer from the school]. George: We had no prior personal experiences from specialised music classes, but there is a lot of music in our home.
[…] We also believe in developing creativity as a general way of learning.
[…] We were also affected by the fact that our nearby school had grown from 75 to 750 pupils. Hilary: Like so many others my daughter sang before she talked […], so music has always been there. […] I had also heard good things about the specialised music classes, and about the school itself. So the quality of the school and my child's interest [played a role in the school choice].
These conversations display the parents' dispositions vis-a-vis the educational market and their actions to implicitly transmit the family's habitus. It reflects their concerns about placing their children into good schools. Several families also sent applications to multiple schools with a specialisation in music, indicating how desirable it was for them to get their children into a specialised educational track. For them, exercising an active school choice was a natural part of parenting.
In contrast to this group, parents in working-class and disadvantaged neighbourhoods reported that the application to the music programme of the local school was their child's own doing. Not only did children in those families play a key role in making a decision, but in the case of the D school, some parents did not even know their child had submitted an application.

Transcript 2 (school D)
Hank: In our case it was our daughter who wanted to attend a music class. She told me that she took an audition. I had no idea. That was how it was. She was the one who took it upon herself. Mohammad: It was my daughter, too; she wanted to do it. I had no idea about music classes, or that there even was such a programme.
In most cases, however, the school choice was described as having been based on leisurely negotiations and, ultimately, agreement between the parents' view and the child's interest.
A striking series of narratives described applying to a music programme in order to avoid the default school, or to enter a secluded space within an otherwise dysfunctional school environment: Transcript 3 (school F) Ibrahim: We had heard that the music classes had order and discipline. My children study at the [names the school], which has the poor reputation of being very noisy, but the music classes are separate and there was supposed to be order and discipline there. We were building our house in the neighbourhood and so became associated with the local school, which I was not so happy about in the beginning. But with the opportunity [to apply to the specialised programme] I felt calmer.
Ibrahim's story resonated with how Lisa (school E) expressed her deliberations: 'I had heard from several others that if you are going to study at this school, you should attend the music class. ' Thus, the framing of the music class was supposed to imply boundaries against the presence of negative influences on schooling.
Many parents associated specialised music programmes with a favourable educational environment. While parents of children in the high-performing schools talked about music education boosting academic achievement in general -'it transfers to other school subjects' (Nancy, School C) -and enriching school experiences, parents at low-performing schools felt that the music programme could provide a breathing space for children with school difficulties, or a refuge within a chaotic environment.
A general view was that the music specialisation establishes strong social bonds between students, and giving them a sense of belonging improves social cohesion. There is value in the collective identity of being a part of a specialised music class. The many extracurricular activities that are carried out by those in music classes teach students how to interact appropriately and take responsibility for joint commitments and arrangements, such as during concerts and traveling.

Transcript 6 (school D)
Kirsten: There is cohesion in the music class. They all have to cooperate. They will be placed in an ensemble where you cannot mess with each other and not make trouble: [they learn] there are boundaries. Jennifer: There are also links between the different grades. My daughter is in the 6 th grade and she feels safe with the [music class] students in 8 th and 9 th grades as well, which I think is pretty unusual in our school [other parents agree].
Students are thought to acquire transferable habits and social codes from the hidden curriculum. Rose-Mary described the parents' request for a stronger classification between the specialised programme students and the regular classes: Transcript 5 (school F) Rose-Mary: I think maybe there is an expectation I have noticed in my son's class that is reinforced in some way: you are expected to be more well-behaved in concert situations when you have to sing together and then bow. There is something useful in participating in this kind of exercise day after day, year after year. At the latest parent meeting, I noticed a common desire that the music students would not be mixed in with the regular classes, but should be kept as a separate group to an even greater extent.
The wish that the music classes remain separated from the regular ones was hinted at in several interviews. It was mainly about preserving the characteristics of the separate track. The parent narratives reflect strategies of building a peer network that enables the transmission of cultural capital, and may itself constitute social capital. Such social ties are also associated with the creation of a common economic capital. Music students earn money from the gate receipts at their public concerts and by charging a fee when they are engaged to perform. Parents contribute by having sales during concerts or apart from the school's musical events. In this way, economic resources are built up that can be used for travel expenses, to support the school's provision of the specialised programme, and to finance policy initiatives.

Differences in the architecture and functionality of the associations
Parents associations operate in ways determined by the interplay between different family and institutional habitus, that is, according to what its members consider acceptable, expected, or desirable patterns of behaviour. Thus, socially and culturally diverse environments result in different behaviours in the interactions with teachers and the school administration.
While parents in working-class environments generally let the school establish the rules (e.g. schools D and E), associations comprised of middle-class characteristics are more disposed to negotiate and take an active part in policy decisions (e.g. schools A and G). Such differences are interrelated with the school's institutional architecture. In school contexts where parent associations have been historically prominent and powerful, the parental network appears to be both operational and strategic. Parents volunteer a substantial amount of their time to organise extracurricular activities, and routinely engage in policy issues. Newer associations, by contrast, are almost entirely operationally focused, devoting their time to organising after-school activities. The strategic work carried out by stronger, more established associations may be understood as the gathering and recontextualisation of capital, whereby parents convert their cultural, social, and economic capital into a collective resource that will support their actions. Well-educated parents know the 'rules of the game' and are more likely to make their voices heard during negotiations on policy. Thus, specialised music programmes often have weakened boundaries between parenting and schooling, where the organised parent interest group, the parents association, represents an intermediate third sector between the public school and the nuclear family unit.
Transcript 7 (school G) Susan: The music classes are a municipal concern, but we stand in a parallel position, because if we do not exert our influence, nothing happens. […] We make decisions and see to it that things get done for the sake of our children. Norman: Politically, we are a single-issue party. We look after the best interests of our children's music classes. That is our only issue, and we actively pursue it in our relations with politicians and other groups. The school administration needs to take a more holistic approach towards the music classes and the regular classes. They may be bound by a principle of equality, but we are not bound in such a way.
The parents whose children attended school G believe themselves to be the driving force behind the music programme. Moreover, they have found a way of engaging in policy dialogue with the school administration in order to ensure that their special interest is served.
During the interviews, the stronger parents associations spoke of strategies for maintaining their power and defended their desire to maintain control over the organising principles of the music programme and preserve its relative autonomy. The parents of school A described how they repeatedly intervened when the municipal school board wanted to narrow the school's broad catchment area. As the parents association believed such an action would reduce competition and consequently result in lower performance among students in the programme, the parents opposed it and successfully maintained the selective status of the specialised curriculum. The following transcript recalls witnessing irregularities in the admission procedure and mobilising in several policy issues and matters of school management.

Transcript 8 (school A) Cesar: We have the capability to go further [than others] as non-profit organisation. […]
There is not much that prevents us from pursuing issues based on our interests. David: I believe there is a greater proportion of well-educated parents who put their children into this type of programme, and who are more familiar with the outcomes. The administration ran into difficulty when they tried to make changes to the school selection and admission procedures because the parent of one of the affected students was a lawyer. This is not the first time something like this has happened. We have educated parents in our association! As exemplified above, middle-class parents associations draw on their professional knowledge and such bureaucratic assets as legal competence and familiarity with public management to advance their personal interests. Thus, parents may take advantage of their cultural capital to exert influence over policy decisions. Our interviews showed that stronger associations also use their financial resources to intervene in specific matters of management, such as decisions to invest in musical equipment.
The differences between stronger and weaker parents associations are usually reflected in their historical structure. Stronger associations have standing committees and frequent board meetings. By contrast, parents associations that interact with school boards infrequently have a comparatively loose organisation and only meet among a few times a year. The out-of-school events they help arrange are usually predetermined by music teachers. Although all parents associations are formally organised (e.g. they are bound by statutes and have annual meetings), the rules of action that govern them include more or less unspoken expectations of parental involvement in the music programme's extracurricular activities.
Transcript 9 (school D) Nathalie: Some parents do not spend any time on the activities. Part of their explanation is that they are not used to participating in voluntary associations. But if you ask for help with baking buns, they will help -and are happy to help. Paul: We have an organisation, but its work is not well defined. You might think the board will arrange things, but it still depends on the parents' involvement. Kirsten: I think some believe the school is the problem. Parents have difficulties seeing the difference, where the boundaries are.
In an oral contract, parents undertake the responsibility of supporting the organised out-ofschool activities of the music programme. However, a problem arises when the invisible code -the rules of the game -has not been understood. Thus, to show initiative and perhaps even become a member of the board presupposes a particular habitus. The lack of such a disposition among parents seemed most prominent in less privileged school environments.

Discussion
For some parents, choosing a specialised music programme may be about doing 'concerted cultivation' , while in instances where a child's own interest guides the educational choice, parents may be more inclined toward favouring 'natural growth' (Lareau 2011). In the latter sense, the children do not generally make strategic decisions on the same basis that middle-class parents typically do (Ball 2003;Reay and Ball 1998;Reay, Crozier, and Clayton 2010;Vincent, Braun, and Ball 2010). For children, finding an accessible social space within the local school -a modest specialised programme -may be enough to spark their interest in music.
' Average' middle-class parents or mixed groups of parents may represent a variety of parenting styles (Bennett, Lutz, and Jayaram 2012). The theory of specialised programmes as a category of school choice options for the informed middle-class requires differentiation with regard to social class factions and the variety of social mixes (Gaztambide-Fernández and Parekh 2017). Any nuanced description of class-based differences should take into account the fact that families may relate to schools having various institutional habitus with different notions of classification and framing Bernstein 1975Bernstein , 2000Power et al. 2003;Power and Whitty 2002;Reay, David, and Ball 2001).
The concept of specialisation in Sweden ranges from elitism to social interventionism. It is a disparate mix that is discursively comparable with the corresponding phenomenon in Canada and the US (Gaztambide-Fernández, Nicholls, and Arráiz-Matute 2016). The great variety of specialised music programmes results in a certain ambiguity in the meaning of specialisation. As the present study demonstrates, most parents opting for specialised music classes for their children had never pursued preparatory training for a music career themselves; rather, they sought for their children the cultivation and socialisation that artistic specialisation was supposed to bestow. In more well-off areas, specialisation may be considered a way of enriching a child's educational experiences, whereas in deprived areas parents envisaged the specialised track as a refuge that might shield their child from an otherwise risky school environment (Gaztambide-Fernández and Rivière 2019; Goldring and Hausman 1999;Vincent, Braun, and Ball 2010). It is primarily from these points of view that the specialised track is perceived as a morally preferable alternative on the educational marketplace (Saifer and Gaztambide-Fernández 2017). In either case, however, parents were attracted by a distinction between 'music students' and 'students in regular classes' that was supposed to be firmly sustained. The underlying principle of specialisation was seen to be that of a strongly framed pedagogic practice through which desirable socialisation may be realised (Bernstein 1975).
While the present study cannot confirm whether middle-class parents are generally more involved in extracurricular activities than their lower-class counterparts, that view concurs with suggestions that parental networks having middle-class characteristics seem to have a greater ability to use symbolic resources in policy negotiation to customise their children's educational experiences (Horvat, Weininger, and Lareau 2003;Lareau and Lopez Muñoz 2012;Weininger and Lareau 2003). In well-established middle-class parents associations, members keep abreast of what is happening in policy processes at different levels in and beyond their child's school. They also seem disposed to fight proposals and decisions that undermine their views on how specialised programmes should be run. They not only know the rules of the game, but they also possess the ability to use them in the policy making arena (Larea, Evans, and Yee 2016). In such negotiations, a strong parents association has the capability of allocating personal resources and amassing a collective force against politicians and the school administration, resources including professional knowledge and bureaucratic assets (Savage et al. 1992). For example, a civil servant's knowledge of laws and regulations may be used to intervene in the parents-school partnership. A parents association can also use its financial resources in interactions with local school management. In this way, confident associations initiate actions on the basis of a relatively weak classification between the contexts of home and school (Bernstein 2000;Horvat, Weininger, and Lareau 2003). By contrast, those parents associations embedded in working-class areas or poor neighbourhoods appear to be unfamiliar with policy issues and lack the experience of having participated in public hearings opposing school boards. Such less privileged and 'weak' associations are more likely to give their unqualified support to the school in any dispute concerning extracurricular activities.
Our findings suggest the need for interrelating the collective behaviour of parents with the culture or 'code' of the particular school. The stronger parents associations are by far those composed of middle-or upper-middle class parents. However, their strength must be understood as embedded in social structures historically rooted in an institutionalised home-school partnership. Parents who constitute effective associations have predetermined entitlements that energize their control and the propel association's relative autonomy, underscoring the contention that home-school interactions are a reflection of the interplay between family and institutional habitus.