Parents' involvement in care order decisions: a cross-country study of front-line practice

This article examines parents ’ involvement in care order decision-making in four countries at one particular point in the care order process, namely, when the child protection worker discusses with the parents his or her considerations regarding child removal. The countries represent different child welfare systems with Norway and Finland categorized as ‘ family service systems ’ and the USA as a ‘ child protection system ’ , with England somewhere in between. The focus is on whether the forms and intensity of involvement are different in these four countries and whether the system orientation towards family services or child protection in ﬂ uences practice in the social welfare agencies with parents. Involvement is studied in terms of providing information to parents, collecting information from parents and ensuring inclusion in the decision-making processes. A vignette method is employed in a survey with 768 responses from child protection workers in four countries. The findings do not show a consistent pattern of difference regarding parental involvement in care order preparations that align with the type of child welfare system in which staff work. The goal in each child welfare system is to include parents, but the precise ways in which it is carried out (or not) vary. Methodological suggestions are given for further studies.


INTRODUCTION
Care orders are state interventions in the private relations of children and their parents. They are based on the assumption that in some situations, parental care of a child should be replaced by public care; therefore, parents' rights to raise a child are restricted or even terminated, and the child is (usually) removed from the home. The final decisions about compulsory 'care orders', which may be termed differently across the countries, are made by courts or court-like institutions in most countries (Burns et al., forthcoming). However, it is the child protection worker's judgement that initiates the decision-making process to remove the child, and in this article, we explore care order decision-making from the point of view of front-line practice. We are especially interested in how parents may be involved in social work decision-making about care orders. We focus on one particular point in the care order process: when the child protection worker discusses with the parents his or her considerations regarding child removal.
Our analysis is informed by the political discourse of Habermas (1996;Eriksen & Weigård, 2004) and his framework for understanding the legitimacy of statelevel decision-making vis-à-vis the citizenry (cf. Rothstein 2011). In this perspective, legitimacy is made more robust when decisions are deliberated with the inclusion of those whose lives are directly affected by state intervention. In the context of child welfare, this means children and parents should be involved in decisionmaking to the extent that they are heard, that their perspectives and interests are included and considered and that they are given adequate information (Authors' own a). To study this empirically and to be able to Child and Family Social Work 2016 compare findings across national contexts, we have used a vignette method so that when workers are asked about parent's participation, they depart from the same scenario.
In the following, we explore parents' involvement in care order decision-making in four countries: Finland, Norway, England and the USA (California). The countries represent different child welfare systems with Norway and Finland categorized as 'family service systems' and the USA as a 'child protection system', with England somewhere in between, a hybrid system, where family services may be provided as part of a highly regulated child protection system (Gilbert et al. 2011). The formal regulations, thresholds and practices for care order decision-making differ in these countries (Authors' own a). We look here at the involvement of parents from the point of view of the representatives of the child welfare systemchild protection workersin the social welfare agencies. We refer to child protection workers as a common term across countries, even though each country may use different terms to describe front-line workers in their child welfare system. The term care order preparations is used to refer to the agency-based policies and practices that help determine whether and when to make an application to the court. We are interested in whether the forms and intensity of involvement are different in these four countries and whether the system orientation towards family services or child protection influences practice with parents.

PARENTAL INVOLVEMENT IN DECISION-MAKING: SKILLS, ATTITUDES AND ORGANIZATIONAL FACTORS
When care order decisions are proposed and made, parents are indeed in a difficult position. Their parental skills and customs to look after their children are contested, and the restriction of their parental rights may take place against their own view of the quality of their parenting. Further, the very examination of the family relations may threaten the rights of privacy in the family. The European Convention on Human Rights, for example, states that everyone has the right to respect for private and family life and this right should be interfered with only when necessary, such as to protect the rights and freedoms of others (Article 8). This statement underlines the private and specific nature of family relations that care order preparations may threaten. The rights for privacy in a family must, however, be balanced against the rights of the child for protection from neglect and abuse. Despite the highly complex and sensitive nature of care order decisions, existing research, in general, suggests that parents' input in decision-making is not a common component of child welfare services (e.g. Dumbrill 2006;Freymond 2007;Klease 2008;Crea and Berzin 2009;Hall and Slembrouck 2011;Höjer 2011;Lietz 2011;Virokannas 2011;Gladstone et al. 2014). When parents have been interviewed about their experiences and sentiments about the child welfare system, they often report frustration and dissatisfaction at being excluded from the decision-making process. The involvement of fathers is a particular challenge for child welfare practice as, often, either they are absent from practice or practitioners do not involve them in the same way as mothers (e.g. Maxwell 2013;Skramstad and Skivenes 2015). Research suggests, however, that birth parents who are engaged in the decision-making process are more likely to participate in services (Littell 2001;Dawson and Berry 2002;Kemp et al. 2009, McLendon et al., 2012. Höjer (2011) found that parents who had positive experiences with child welfare had been invited to participate in decision-making, while many of those with negative experiences had not.
The reasons for poor involvement of parents are, in short, threefold. First, it may be a consequence of the skills, values, techniques and attitudes of the practitioners involved (e.g. Gladstone et al. 2014). Even if formal regulations require parental involvement, front-line workers and managers will often have the authority to exercise discretion in their day-to-day work and thus determine whether and how parents are involved, or workers use unauthorized discretion in their handling of parents. Research on practice finds variation in the way that child welfare workers engage parents in decision-making (Kapp and Propp 2002;Estefan et al., 2012;Gladstone et al. 2014). When caseworkers involve parents in decision-making, they do so to varying degrees; according to parents' reports, the variation is by caseworker rather than by agency or by state (Lietz 2011;Farrell et al., 2012). According to Smith and Donovan (2003), caseworkers too are conscious of this dynamic.
Secondly, some parents are unable or unwilling to engage constructively with child welfare services. They may have been badly damaged by their own experiences or have limited social and cognitive abilities; therefore, it is not a question of blaming them for avoiding or resisting, but being realistic about their capacity and interests. In reviewing investigations into cases in England where children have been seriously abused or neglected, Brandon et al. (2009) note frequent lack of parental cooperation with agencies and list the forms this can take: 'deliberate deception, disguised compliance and "telling workers what they want to hear", selective engagement, and sporadic, passive or desultory compliance' (Brandon et al. 2009: 3, 76). They conclude that workers have to be careful in case 'efforts not to be judgemental become a failure to exercise professional judgement' (Brandon et al. 2009: 27).
Thirdly, the literature suggests that there are structural reasons for the poor involvement of parents. This could be due to the organizational routines and assumptions that emphasise the professional view over the parents' view. Research shows the ambiguities and limitations of parental involvement and representation in the meetings in which decisions should be made. For example, studying involvement of parents in child protection case conferences in England in the 1990s, Corby et al. (1996) concluded that parental attendance was not meant to bring different decisions or different outcomes, but to bring about parental compliance with the predetermined views of the professionals. Dickens et al. (2015) observed the same tendency in 'edge of care' meetings in England in 2010-2011. As regards the USA, Alpert (2005) suggests that barriers for parental involvement derive from the intrinsically punitive structure of child welfare under which caseworkers are trained to view children as their primary clients and child protection (usually protection from parents) as their primary goal, creating institutional bias against birth parents if the risk threshold for the child is reached. Given that the risk threshold for removal is arguably high in the USA, we might find that these institutional arguments hold more firmly in 'edge of care' or 'in care' cases than they might be true for lower-risk cases. And whether this argument remains true in all jurisdictions that are increasingly adopting family-centred approaches to practice is also unclear (National Resource Center for Family-Centered Practice .) According to Corby et al. (1996), it is important to recognize the conflicts of interest and motivation between parents, children and practitioners and to be realistic about the extent to which all parents can and want to be actively involved in making decisions about the care of their child. Parental participation may easily turn into rhetoric if the conflict of interests is not recognized in this particularly sensitive and power-rich decisionmaking arena. Even when the parents are genuinely, ethically and effectively involved in the decision-making process, the decision may be against their view.
In addition, parental involvement in care order decision-making is influenced by the legislation, regulations and rationales of the child welfare systems. Those factors set certain parameters for child protection practitioners about how and when to involve parents, and they define the rights of the parents in question. The organizational and institutional contexts for care order decision-making vary across countries. This variation will be next explored in the four countries under study as it is argued that macro elements such as the countries' policies should be explored in order to make sense of front-line decision-making (Smith and Donovan 2003;Duffy and Collins 2010).

THE INSTITUTIONAL FRAMES FOR PARENTS' INVOLVEMENT IN CARE ORDER DECISION-MAKING IN CALIFORNIA, ENGLAND, FINLAND AND NORWAY
The common principle shared by the countries studied here is that it is the parents' right and duty to look after their children and that the accomplishment of these duties and rights is ultimately regulated by the state (Archard 2003;Gillies 2008). They also share the principle that when concerns arise about the parents' care of their children and the child welfare system becomes involved, the child's situation should be carefully examined, and the parents should be involved in the process (Authors' own a). There are, however, differences in defining who 'the parent' is and how and when to involve her or him or them.
In California, the birth mother is typically considered the child's parent (although a presumed mother or adoptive mother may be involved in some circumstances). The paternal parent is either a biological father, an alleged father, an adoptive father or a presumed father (Cantil-Sakauye et al. 2011). In England, parents are biological (or adoptive) parents. Alongside the notion of parentage, there is the legally important notion of 'parental responsibility'. In English law, a birth father will not necessarily have parental responsibility, and it is possible for others to acquire parental responsibility and share it with the parents. People with parental responsibility (not only parents) will be parties to care order proceedings. In Norway, usually, parents are the biological parents, if they are married, are cohabiting or have made an agreement about joint responsibilities. If not, the biological mother alone has parental responsibility. Custodians are those who have the daily responsibility for the child, and a care order will only restrict the custodian's responsibilities. As for involvement in care order proceedings, the Norwegian legislation uses the term 'party in a case' who is a 'person to whom a decision is directed or the case otherwise directly concerns'. This means that in a child protection case, it is primarily the parents with custody rights and the child himself or herself when 15+ years who are parties. In Finland, 'parents' and 'custodians' have a different position in the process. Parents are normally custodians, but parents may also lose custodial rights, and they can be given to other persons. The custodians should be involved in every part of the process, and their views have an impact on care order preparations (e.g. they may contest the care order). Those parents without custodial rights have fewer legal rights than the custodians, but they have, for example, the right to make an appeal against the care order (Pösö and Huhtanen, forthcoming).
In all countries under study, the care order decision restricts parents' custodial rights. Parental rights may be also terminated in England, California and Norway (although it happens in a relatively small proportion of cases in Norway). In Finland, adoption is not a placement option granted by the Child Welfare Act, 2007.
Legislation and other statutory regulations in the countries studied here include statements about the involvement of parents. In California, the parents have rights to notification, counsel and reunification services. The parents must be interviewed during the investigation. Some California counties use Team Decision Making meetings or Family Group Conferences to include parents, other significant adults and older youth to devise a safety plan for the family. Parents also have rights to a legal representative to ensure their rights throughout the child welfare process (Cantil-Sakauye et al., 2011) In England, involving parents throughout care proceedings is required by primary legislation, formal regulations and statutory guidance (Children Act 1989; DfE (Department for Education) 2014). The parents are entitled to have their own lawyer, paid for out of government funds (legal aid), and she or he will advise and represent them in pre-proceedings processes and the court proceedings. Social workers are required to continue to work with the parents throughout. There are legal obligations that parents' wishes and feelings are ascertained and taken into consideration at all stages of the process (even if, ultimately, they are not followed). In Norway, the agency has an independent obligation to inform parents about all legal regulations and about parental rights and obligations, according to the Administration Act and the Child Welfare Act, 1992. It is underscored that parents must be informed about the opportunity to have a representative or a lawyer; in practice, parents are entitled to have a lawyer, paid for out of government funds from the day they are informed that the agency has decided or is seriously considering presenting the case to the county board. Similar to Norway, the Finnish legislation emphasizes information: parents should be informed, and they should have access to all the material in the care order preparations. In addition, the Finnish legislation underlines that the parents' and custodians' views on the care order should be carefully considered. There is a special occasion called a 'hearing' in the social welfare agency, including at least the custodians, the child and social workers, in which the custodians' and child's views are considered. Legal representatives may attend the hearing as well, but their presence is not required. The views of the custodians and children presented in the hearing have a fundamental impact on the further decisionmaking process: if the custodians or a child who is 12 years old or older disagrees with the care order proposal, the decision will be made by the administrative court (involuntary care order); if they agree with it, the decision will be made by a local authority (voluntary care order). The custodians as well as non-custodial parents should be involved in the care order proceedings also before and after the hearing. There is no regulation that legal assistance should be provided in the care order preparations in the social welfare agency.
In sum, parents' involvement in care order preparations is a shared feature of the statutory regulations of the child welfare systems studied, despite the wider sociopolitical differences between the countries. The detailed arrangements vary, notably regarding the involvement of lawyers, but on the face of it, the general principles are similar: that parents should, as far as possible, be informed, consulted and involved in the decision-making process. Given these shared principles, we anticipate that child protection workers will generally rate the involvement of parents as high, but our study was designed to investigate whether there were cross-national differences within that overall, formal commitment and, if so, how these could be understood.

METHOD AND DATA
The analysis of the involvement of parents in decisionmaking about care order proceedings in California, England, Finland and Norway is based on child protection workers' responses to the following vignette presented in an on-line survey: You are working with a boy -Alexwho is 5 years old and whose family has received in-home services over a period of time. The case includes parental substance abuse, previous domestic violence, and general neglect. The circumstances of the case have deteriorated recently to such an extent that you are concerned that the boy's risk of harm is high. You are starting preparations for care order proceedings with a view to removing Alex from his parents, and you have an interview with the parents to inform them about this. The parents are opposing a removal of Alex.
Respondents were asked to respond to the vignette based upon the information provided. This hypothetical case description was designed to present staff with a situation that might be familiar in child protection practice in each country. Although the nature of child welfare agencies' responses to neglect and substance abuse may vary across the countries because of different thresholds for intervention (Authors' own a), the vignette suggests that the circumstances may now require an intensive intervention. We also acknowledged that this interaction may be the first or among the first encounters between a worker and a parent in California, whereas such a meeting would, by design, typically follow a long-standing engagement in the other three countries. We emphasise, however, that the child protection worker would usually interact directly with the parent(s), and thus, the focus on this interaction addresses cross-cultural comparability (e.g. Soydan 1996) in our design.
We asked child protection workers about their aims in the meeting with parents. They were asked to use a scale between 1 and 5 to mark how important the aims were under the themes of (1) providing information to the parents about care order preparations, (2) collecting information regarding care order-related tasks and (3) ensuring the inclusion of parents in the decision-making process. On the theme of providing information to the parents, specified aims for the conversation were 'Inform the parents about why you are starting care order preparations', 'Confirm that the parents understand that care order preparations are in process' and 'Advise the parents to seek independent advice from a lawyer'. Under the theme of collecting information from parents, the detailed aims were 'Establish the facts and circumstances of the casewhat really happened', 'Find out if the parents have any needs that you can help with', 'Ask for parental preferences about potential future placement of the child' and 'Make sure parental views about the needs of the child are heard'. The third theme was captured in the aim 'Make sure the parents are included as participants in relevant decision-making processes'.
The vignette was included in an on-line survey as part of a research project about decision-making in child welfare. It was addressed to front-line child protection workers via contacts in relevant agencies and social work associations. The survey was answered from February to June 2014. The data gathering process, including research ethics, is described in detail at the website http://www.uib.no/admorg/85747/survey-material#social-worker-survey. A total of 768 eligible responses were received from child protection workers who are involved in care order decision-making. The total numbers of respondents from each country were 370 from Norway, 206 from Finland, 103 from England and 89 from the USA.
The typical respondent had worked in child protection for 5.9 years. Respondents from England were, on average, younger (between 25 and35 years) than their colleagues in Finland, Norway and the USA (between 36 and 45 years). Respondents from California were more highly educated than the other workers; 91% possessed an Master of Social Work degree. In Finland, 65% had an MA degree (suggesting that upwards of 35% of respondents in Finland were unqualified temporary workers because they lacked the MA degree), in England, 57% had a Master's degree and in Norway, 9% had a MA degree. In all four countries, the vast majority of workers sampled were female.
We used the statistical programme STATA and conducted simple correlation analyses, chi-squared tests and mean comparison t-tests. The scores were added together and divided by the number of respondents to give a mean score for each aim in each country and the overall (inter-country) means. Table A in the Appendix presents the mean comparison between countries, one-way analysis of variance tests, with reported mean difference. Significance is displayed as follows: ***P < 0.01, ** = P < 0.05, * = P < 0.1, with the awareness that the P < 0.05 and P < 0.1 are on the margin of what is relevant to report as statistically significant. The main results in the findings section are presented in graphs, using bars displaying the mean score for each country on the variable in question, with the confidence interval of 95% indicated on the brackets on each mean score. We first ran an analysis of variance test in STATA, followed by a margins test, setting a confidence level of 95% and the 'country' variable as identifying the margins. We were thus provided with a graph displaying the mean score and confidence levels for each country on the variables.

The involvement of parents as an important aim for conversations
The overall finding is that parental involvement is an important aim for child protection workers in all four countries when they are confronted with a situation in which they are starting care order preparations. Child protection workers in the four countries regard the various aims of involvement as being either important or very important (mean values between 4.1. and 4.8 on all dimensions). In particular, the aims of informing the parents why the care order preparations are being taken, and ensuring that they understand this, are emphasized as important in the survey responses. Reviewing the overall scores, the aims of advising the parents about seeking legal advice, collecting information about the parents' needs, preferences about the child's possible placement and what really happened were among the least important, but as Table 1 shows, there was considerable variation between the countries on these issues, which we explore further later.

Types of conversations with the parents: the differences among the countries
In all of the countries, child protection workers strongly aim to inform the parents about the reasons for the care order proceedings and to ensure that the parents understand that the care order preparations are in process (Fig. 1a-c). These aims are seen as being even more important by the Norwegian and Finnish child protection workers than their colleagues in England and California (P < 0.01). In the Appendix, Table A, an overview of calculations of significance differences is presented. The pattern is, however, different as regards informing the parents about seeking legal advice: now, the Norwegian and English child protection workers are similar, and they view this as an important or very important aim more often than their Finnish and California colleagues (P < 0.01); California responses are the lowest (P < 0.05).
As regards the theme of collecting information from the parents (Fig. 2a-d), the child protection workers in California score the highest in every studied dimension. In particular, they regard the aim of finding out what really happened as being more important than their colleagues in other countries (P < 0.01). In this dimension, the Finnish and Norwegian responses are similar, and they differ from the English responses (P < 0.05). When aiming to collect information about the parents' needs, the English, Finnish and Norwegian responses are similar and lower than the responses from California (P < 0.05). When aiming to learn about the parents' preferences about the potential future placement of the child, again, the California child protection workers put this as a more important aim than their colleagues. The difference with the Norwegian responses is largest (P < 0.01). When aiming to hear the parents' views about the child's needs, the English child protection workers score the lowest. The difference between the California and English responses is significant (P < 0.01), whereas there are no significant differences between the Finnish and Norwegian responses.
The aim to ensure that the parents are included as participants in the care order preparations is regarded highest (and similarly) by the California and Finnish child protection workers and lowest (and similarly) by the English and Norwegian respondents (P < 0.01) (Fig. 3).

DISCUSSION
The data clearly indicate that involvement of parents is important in all four countries, suggesting that this is a highly regarded norm in child welfare practiceor a common form of rhetoric, as suggested by Corby et al. (1996)regardless of the country's statutory frames. The cross-country differences presented in this article reflect the relative importance between countries of involving parents. In the scenario presented in the survey, child protection workers prioritized the aims of informing the parents about the proceedings, whereas the other aims were regarded as important, but relatively less so. The survey responses echo the legislative and policy frames in these countries, as well as the orientations of the child welfare systems. As discussed earlier, the California legislative frame provides parents with indirect access to required information via lawyers, a practice that resembles police investigations (Dumbrill 2006). Front-line practitioners' responses may reflect this policy frame as workers there were less inclinedcompared with their international peersto indicate the high importance of giving information to parents. This does not, of course, mean that child protection workers in California find the provision of information an unimportant task, but they may see that this activity is delegated largely to attorneys. Another interpretation is that in a situation such as that presented in the vignette, the worker's main focus is to extract information from the parents in order to assess the child's safety and risk of harm; other conversations with the parent may not serve as centrally important. The interaction described in the vignette would also typically take place during the 'investigation', a term that denotes information and evidence gathering. The Nordic countries in particular were strong in aiming to give information to the parents about their justification for a care order and ensuring the parents' understanding. These responses may reflect their family service system orientation, in which a therapeutic approach requires that parents are informed so that they can make changes in their patterns and behaviours. Having enough information about the proceedings is a fundamental precondition for the parents to exercise their rights, whether or not they choose to work in partnership with the child welfare agency (cf. Ife, 2012). The values practitioners place on informing parents about seeking legal advice speak to the complexities of studying child protection frameworks that are support oriented vs. protection based. Finland and California, for example, both score relatively low on this dimension, but the reasons are very different. In California, workers' responses are most likely due to the prescribed procedures for care order proceedings: all parents are routinely assigned a lawyer (if they do not secure one themselves), as such, it is not a priority to advise parents to do so. Legal advocacy for parents is the norm in care proceedings in Norway and England as well, but child protection workers aimed to inform the parents about this in any case. In England, although the lawyer is paid out of government funds, parents must seek out the lawyer's services (they will not be assigned one unless they dothe vast majority do). In Norway, a shift to care order proceedings is a significant change and marks the transition from voluntary to involuntary intervention. Thus, informing parents about their legal rights is vital. All this is in marked contrast to Finland, where care order preparations in the social welfare agency are not commonly seen in terms of juridical processesmost of them do not end up in court as they are decided as 'voluntary care orders' by the local authorityand therefore, advice on legal advocacy is perhaps not seen as especially important. This view has recently been challenged as processes resulting in any type of care order decisions indeed have legal consequences for the family. Some have argued that the juridical element should be considered at this stage regardless of the type of the decision (voluntary or involuntary;e.g. Huhtanen 2001;De Godzinsky 2012).
In short, the mixed findings regarding practitioners' views about directing parents to legal advice do not necessarily follow the protection-based vs. family supportbased child welfare states in which these respondents are nested. Norway has developed a rich family support system, yet the role of lawyers is nevertheless important; in Finland, this is not the case. In the English system, parents may be advantaged when represented by a competent lawyer who can clarify where and how family support services could be improved, without the need for their child to be removed (e.g. Pearce et al. 2011). And in California, lawyers are appointed as a matter of course, relieving child protection workers from the task of recommending this.
The California child protection workers, instead, place a higher priority than colleagues in other countries on collecting information from parents. This is likely because in the other three countries, this information was  obtained at an earlier stage in the worker's interactions with the family. Well before a care order application is considered, parents would have received significant child welfare services. In California, workers may not have been involved with the family previously or for any great duration. This results in the same principle being followed across all four countries: that parents should be heard during the care order proceedings, but it also reflects the distinctive organizational approaches in each of these four systems. In California, preparations for care order proceedings are much more time limited than in the other three countries (Authors' own a). Therefore, information gathering is intense at this stage of the process. The differences among the child protection workers' views about including parents as participants in decision-making are intriguing: again, the Finnish and California practitioners give similar scores, both apparently valuing parental involvement more than their colleagues in Norway and England. The responses from California are probably best explained by the strong emphasis on practice improvement in that state in recent years. Innovations in adopting family-centred practice paradigms have taken hold in many California jurisdictions, with many child protection staff working diligently to adopt a philosophy of practice that is inclusive of parents (Child Welfare Information Gateway, n.d). That state has developed a series of 'core practice models' to which child welfare workers and students of social work are being trained and supported through various state and regional training academies (CalSWEC (California Social Work Education Center) 2015). These models provide behavioural illustrations to staff suggesting exemplary strategies for working with children and parents. It is important to note that the large majority of study participants from California possessed a Master's degree in Social Work (or a related field), where they would likely be exposed to ideal models of practice; all of the counties participating in this study also offer ongoing in-service training opportunities for staff. In Norway and England, the lengthy engagement with families along with the availability of legal advice may be the reason workers do not stress parent participation in this particular meeting. The Finnish workers' responses reflect the strong role of the parents in the care order preparations, and therefore, child protection workers take it for granted to include them as participants.
In summary, the findings do not show a consistent pattern of difference regarding parental involvement along the lines of the well-known child protection or family service orientations. The ambition in each child welfare system is to include parents, but the precise ways in which it is carried out (or not) vary. The findings show that the overall ambition should not be taken for granted; it is important to look in detail at the way that legislation, procedure and organizational context shape the way that it is understood and practiced.

CONCLUSIONS
Results from this study underscore the value child protection workers ascribe to parents in their practice. Although their views do not align cleanly with the child welfare system contexts in which they are embedded, the investigatory functions of child welfare agencies in California highlight the importance of collecting information from parents, whereas the family support systems of child welfare in the Nordic countries emphasize the value of providing information to parents. The overall consensus about the involvement of parents among child protection staff across the countries can be contrasted with other data from the same survey about the involvement of children (Authors' own b). Here, there was much greater variation in the workers' responses regarding children's involvement in care order preparations. This supports a recent cross-country analysis of child welfare systems by Gilbert et al. (2011), which noted that a child-centric orientation is an emerging trend, although not yet well established. The decision-making systems take adultsparents in this caseas the standard partners, and children are given a more contested and rhetorical position in child welfare decision-making (e.g. Meyer 2007; Thomas 2007). The traditions of viewing the parents as the key actors have a much longer history in child welfare ideology, policy and practice (Fox Harding 1997), which is reflected in these results. Whether and how parents perceive workers' efforts on their behalf cannot be ascertained from this study, but greater efforts to align good child protection intentions with good outcomes for parents, is an area ripe for further investigation.
The central conundrum at the heart of this discussion is that although the legislative requirements and the ambitions of front-line practitioners hold parents' involvement to be centrally important, the existing research, presented earlier in this article, highlights many shortcomings in achieving this. In this respect, it is important to acknowledge the nature of our data: the responses to the case of Alex give the workers' views of front-line practice that is seen as 'good' if not even 'ideal'. This is the strength of a hypothetical vignette as it crystallizes the normative core of the practice in each country. The normative core includes the involvement of parents and thereby aspires to a decision-making process that meets standards of legitimacy. However, there are at least two methodological challenges in studying participation. First is the conceptualization of participation and how to define and operationalize it. A learning point of our study is that in order to better grasp the meaning of participation in child protection, we need to assess each of the three dimensions we have discussed in this article (providing information, collecting information and ensuring inclusion in the decision-making processes) and to recognize their country and systemspecific meanings. The second challenge is the interrelations of participationit is not sufficient that one party believes participation is going on. The research as reviewed earlier in this article tends to address the experiences of parents about involvement. The lived experiences of decision-making are categorically different from the ambitions of the child welfare system and its representatives, as the study by Gladstone et al. (2014) shows. Moving from ambitions to reality, future studies should endeavour to examine child protection workers' authentic efforts to engage, listen to and provide information to vulnerable parents to reveal the scope and depth of true partnership and how the different child welfare systems either support or hinder this.

FUNDING
The research project is funded by the Norwegian Research Council, with some additional support from the Peder Sather fund University of California at Berkeley.