When standardization becomes the lens of professional practice in child welfare services

Correspondence Marina S. Sletten, Faculty of Health and Welfare, Østfold University College, Fredrikstad, Norway. Email: marina.sletten@hiof.no Abstract This paper examines the relationship between professional work and standardization. There has been an increase in the use of standardized programmes in child welfare services (CWS) in Western society. Some researchers have criticized standardized programmes suggesting that they undermine professionals expertise and threaten their position, whereas others argue that such programmes strengthen professional practice. In this paper, we examine how standardized tools, in this case, a standardized parenting programme and a standardized Norwegian assessment tool, influence professional roles as experienced by child welfare workers (CWS professionals) in Norway. Semistructured individual and group interviews were conducted with 31 frontline workers in two CWS agencies. Our findings suggest that standardized tools increase the social workers experienced professional competence but challenge their professional knowledge base, reflective practice, and professional accountability. Professional and practical implications for CWS work are discussed in the light of these findings.


| INTRODUCTION
Internationally, there is a trend to standardizing child welfare services (CWS). A key aim is to improve the quality of practice by strengthening the professionals' knowledge base through scientific methods (Lyneborg & Damgaard, 2019;Noordegraaf, 2015) and to ensure accountability in professional services and decision-making (Timmermans & Berg, 2010;Webb, 2006). Standardized tools characteristically provide forms and guidelines for predecided actions rather than being based on individual judgement (White, Hall, & Peckover, 2008). As such, they are "procedural standards" that "prescribe the behaviour of professionals" (Timmermans & Berg, 2010, p. 26). This is a form of social regulation related to norms. The tools guide practice, and in this way, standardization forms the new normative standards of social work (Brunsson & Jacobsson, 2002;Møller, 2018). As a result, concerns have been raised about professional autonomy, critical reflection, and objectivity (Timmermans & Berg, 2010).
Some scholars claim that this trend deprofessionalizes social work (e.g. Healy, 2009;Munro, 2005;Ponnert & Svensson, 2016;White et al., 2008). Arguments are that standardization oversimplifies practice, affects professionals' skills, and limits their action (Brodkin, 2008;Gillingham, 2011;Webb, 2006). Gillingham and Humphreys (2009) argue that decision-making tools favour the needs of management and undermine development of professional expertise. The argument that standardization makes social work practice more transparent and auditable is problematic because of the complexity that often characterizes social work (Thompson, simplifying the complex aspects of human existence and not meeting the needs of individuals (Munro & Hardie, 2018).
The arguments presented so far suggest that standardization is at odds with traditional professional work. Professionalism consists of specialized abstract knowledge (Abbott, 1988), with strong discretionary space (Lipsky, 2010). In social work, professional knowledge, skills, and values should, when brought together, lead to accountability (Thompson, 2016). CWS professionals deal with complex family situations, and action is often necessary despite uncertainty, ambiguity, and fallibility (Munro, 2005). When professionals deal with complex cases, there is a need to apply sensible, local knowledge, which can be both explicit and tacit (Noordegraaf, 2015). It can be challenging to achieve this quality when standardized tools demand that everything is made explicit (Brunsson & Jacobsson, 2002).
Despite growing concerns about how standardization influences the professional role of social workers, standardization seems to continue to gain ground. With few exceptions (e.g., Gillingham, Harnett, Healy, Lynch, & Tower, 2017;Vis, Lauritzen, & Fossum, 2016), there is limited research on how professionals use and regard standardized tools in their daily work and what implications such tools have for professional roles. The aim of this study is to explore how social workers experience two standardized tools commonly used in Norway and how the tools influence the professional role of CWS workers.

| Context of the study
Norway is characterized by strong egalitarian and redistributive values (Forsberg & Kröger, 2010). The CWS is service-oriented with a child-centric approach, in which the best interest of the child is a core principal. The threshold for early interventions is relatively low, and measures, voluntary or compulsory, aim at preventing risk and promoting a healthy childhood (Skivenes & Søvig, 2017). The majority of CWS measures are thus voluntary inhome services (Statistics Norway, 2019). Hence, the CWS is both protective and supportive in its approach to children that are living under conditions that represent a risk to their health or development, and it can provide a variety of welfare services to improve the living conditions for the child (Skivenes & Søvig, 2017). CWS is regulated through law, which has incorporated the Convention of the Rights of the Child (CRC, 1989).
A variety of standardized tools is used in Norwegian CWS. In this study, we focus on the Kvello Assessment Framework (KF) and the Circle of Security Parenting (COS-P). Both consist of forms and guidelines with predecided actions linked to scientific evidence. In this sense, the expert knowledge is embedded in the rules constituting the standard (Brunsson & Jacobsson, 2002). Neither of the tools are mandated by the state authorities. An ongoing debate in Norway is on whether the use of assessment tools in CWS should be mandated to improve the quality of assessments and achieve more equal practices (Lauritzen, Vis, Havnen, & Fossum, 2017).
KF is a nonlicensed standardized assessment framework developed by a Norwegian psychologist to direct the decision-making process (Kvello, 2015). The framework is used in various forms in over 50% of Norwegian municipalities (Vis et al., 2016) and has been implemented in agreement between the private developer and the local authority (Lauritzen et al., 2017). KF shares similarities with the Swedish Barns Behov i Centrum (Child's Needs in the Centre) and the Danish Integrated Children's System that originate from the British Common Assessment Framework (Lauritzen et al., 2017). The KF has an ecological theoretical underpinning that offers a guideline for how to carry out an "assessment of needs." This includes obtaining information and assessing the needs of the families by using different approaches (e.g., structured parent interview). KF focuses on identifying risk and protective factors, and the structure of the form promotes splitting the information into different sections (e.g., child's self-report and the caregiver's understanding of the child). The professionals fill out the form and address the sections relevant for the particular case.
There is no manual describing KF, and this is a limitation of the tool (Lauritzen et al., 2017). KF recognizes COS-P as an important supplement for assessing the parent-child interaction.
COS-P originated from Circle of Security (COS). COS is an early intervention manual and licence-based programme developed in the United States and rooted in attachment theory. The aim is to train caregivers to develop reflective capacity and enhance their understanding of their child's attachment needs (Powell, Cooper, Hoffman, Marvin, & Zeanah, 2013). Key elements of COS is use of graphics, videos, and various core metaphorical concepts, in which parents are invited to reflect upon fixed questions asked by the therapist. COS was originally developed as a standardized 20-week group intervention programme. In 2013, the 8-week COS-P programme was launched with the components of the COS intervention protocol, along with COS graphics and a DVD protocol. The DVD with archived videotapes aims to promote secure attachment through this step-by-step process for use in both group and individual sessions (Powell et al., 2013). COS-P is designed as a costeffective programme for broad implementation with little training (Cassidy et al., 2017). COS-P differs from COS interventions by excluding a preliminary assessment, not individually tailoring the video, and moving the subject through the components at a faster rate (Pazzagli, Laghezza, Manaresi, Mazzeschi, & Powell, 2014).
Although state authority has recommended and facilitated COS-P for CWS, it has, to a great extent, been implemented of the initiative of individual social workers.
It is important to note that KF and COS-P are used in different phases of the casework process and complement each other. Our aim is not to compare the two tools but, rather, to use them as a way to explore how standardized tools influence the professional role.

| Previous research on standardization in social services
In CWS worldwide, numerous standardized tools have been implemented to improve services for families and children, such as decision-making tools (Gillingham et al., 2017;Vis et al., 2016) and parental intervention programmes (Mercer, 2014;Wike et al., 2014;Yaholkoski, Hurl, & Theule, 2016). This development relates to the ideas of evidence-based practice (EBP) and managerialism. Evidencebased practice gained acceptance because of a need to legitimize professionals' work (Timmermans & Berg, 2010). When standards are linked to science viewed as expert knowledge, the legitimacy increases (Brunsson & Jacobsson, 2002). Furthermore, there has been an amplified dependence on procedures and rules as a response to dealing with risky situations (Webb, 2006). This is to ensure the quality of professional practice (Munro, 2005;Thompson, 2016). Although scholars have raised the concern that standardized tools may undermine professionals (Webb, 2006), such tools have become prominent in the social welfare profession (Healy, 2009).
Research on standardized tools, in particular interventions, has mainly focused on effects (Mercer, 2014;Yaholkoski et al., 2016). Also relevant for this study is research on how standardization influences professional CWS practices (Gillingham et al., 2017;Gillingham & Humphreys, 2009;Léveillé & Chamberland, 2010). However, this research is not consistent.
Research suggests that standardized tools may have a negative affect on social workers' capacity, as such tools are time-consuming and increase the workload (Léveillé & Chamberland, 2010;Vis et al., 2016;Wike et al., 2014). Moreover, some research shows that standardized tools may lead to inaccurate risk assessments (e.g., Benbenishty et al., 2015). Risk factors are challenging to operationalize (Sørensen, 2016;Vis et al., 2016) and may lead to an oversimplification of the family situation (Stanley, 2013). Conversely, studies have shown that standardized tools provide CWS professionals with a language to express their work more accurately (Gillingham et al., 2017;Mercer, 2014;Sørensen, 2016). Furthermore, standardization may lead to more focused and structured CWS work (Almklov, Ulset, & Røyrvik, 2017;Barlow, Fisher, & Jones, 2012;Gillingham et al., 2017) and allow practitioners to generate more information in assessments (Bartelink, van Yperen, & Ingrid, 2015;Léveillé & Chamberland, 2010;Vis et al., 2016). That said, practitioners may struggle to make sense of the amount of information obtained (Barlow et al., 2012).
Research also shows that standardized tools have the potential to strengthen social workers' professional role (Gibbs & Gambrill, 2002;Ponnert & Svensson, 2016;Vis et al., 2016), for instance, by allowing practitioners to become more confident (Almklov et al., 2017;Gillingham et al., 2017;Vis et al., 2016) and gain legitimacy (Ponnert & Svensson, 2016;Stanley, 2013). Some research has found that standardized tools have the potential to foster user involvement when taking a more holistic approach (Bartelink et al., 2015;Léveillé & Chamberland, 2010), whereas other studies revealed it to obstruct user involvement (Almklov et al., 2017;Léveillé & Chamberland, 2010). Research suggests this to be an issue when interventions fit poorly with the given context (Wike et al., 2014).
Studies, which focus on standardization in relation to the use of professional discretion and reflective practice, suggest that standardized tools limit the use of discretion (e.g., Gillingham et al., 2017;Ponnert & Svensson, 2016;Wike et al., 2014). That said, some tools are found to have the flexibility needed to foster professional discretion and critical thinking (Evans, 2010;Gillingham et al., 2017; Høybye-Mortensen, 2013).
Most of these studies are from England and Australia; countries with a more risk-oriented CWS approach compared with Nordic countries. To complement the existing body of research, this study offers an in-depth analysis of how standardized tools influence the professional role within Norwegian CWS.

| Design
To address the research question of how standardized tools influence the professional role, we have developed an exploratory single-case study design (Yin, 2014). Standardization constitutes the case of the study: KF and COS-P constitute exemplifying cases (Bryman, 2016, pp. 60-63). The study has a qualitative design exploring professionals' perspectives on the use of KF and COS-P in everyday CWS situations and is part of a larger PhD project.

| Recruitment and sample
Two CWS offices in different parts of Norway participated in the study. The management team in each office helped facilitate participation by CWS professionals. Office A is a CWS agency with about 90 employees located in a larger city. They have used KF and COS-P for several years, in addition to other standardized tools not included in this study. Although office A constitute the main sample, we also included office B to get a richer data material. Office B has about 30 employees, located in a more rural area. The two offices are organized differently; office A divide tasks into assessment and family services units, whereas office B has a more generalist structure. Furthermore, office B had just recently started to use KF and was in an early phase of using COS-P. Therefore, they would have more recent experiences with implementing the tools. Differences in time of experiences, size, and location could possibly contribute to deeper insights to our research question. However, no significant differences were identified in the two offices; therefore, we will not make distinction between the two when reporting the findings.
Three levels of samples were combined (Bryman, 2016): tools, context, and frontline practitioners. To be included, participating frontline workers had to have experience of using KF and/or COS-P as part of their daily practice and more than 12 months' experience in CWS work. A total of 31 CWS workers (29 women and 2 men) consented to take part in the study. All had a bachelor's degree in social work/child welfare, and some had additional education. They had worked in the CWS from 1 to 20+ years. In office A (n = 26), 17 participants had experience with KF and nine with COS-P. In office B (n = 5), all participants had experience with KF, whereas two had some experience with COS-P.

| Data collection
We used multiple sources of qualitative data collection approach (Yin, 2014). These included 17 semistructured individual interviews (office A: n = 11 and office B: n = 2). Individual interviews allowed participants to share their perspectives and experiences more freely and in depth without boundaries of what others, for example, colleagues, may think. Topics addressed concerned participants' understandings of standardization, experiences with using the tool, reasoning for their actions, and if and how they found the tools supportive for their work.
Two participants were willing to be interviewed twice to provide insights in how they worked in a particular case over time. Additionally, seven unstructured interviews (Bryman, 2016) were conducted over a period of 4 months with three CWS workers from office B. This allowed for insights in how participants from office B gained experience in using the tools by revealing their immediate reflections upon their experiences. Topics in the unstructured interviews comprised of how they used and regarded the tool in their daily practice.
Preliminary findings from the KF interviews revealed different "dimensions of meanings" (Aase & Fossåskaret, 2014). In order to explore these further, seven semistructured group interviews (Bryman, 2016) were arranged with 2-5 participants (office A: n = 17 and office B: n = 2). The purpose of the group interviews was to generate discussions and meta-reflection on participants' experiences, perspectives, and actions (Morgan, 1998, p. 25). To reduce power dynamics within the group, the group was put together with participants holding the same position, working within the same team, and were used to working with each other. For all participants, some background information (education, professional expertise, and working experience) was also collected. All interviews lasted between 60 and 90 min and took place at the CWS work site.

| Data analysis
The multiple source of data generated thick descriptions. Transcripts form the interviews were analysed by applying a data-driven conventional content analysis (Hsieh & Shannon, 2005) to search for themes of common meanings (Krippendorff, 2019) of how standardized tools influence the professional role. The analysis was performed by following the steps of qualitative content analysis described by Graneheim and Lundman (2004) and supported by Nvivo 11. This approach involves moving back and forth between description and interpretation of the transcribed text, searching for both manifest content, such as the participants' explicit statements, and latent content, such as our interpretation of the meaning of the participants' statements. The researchers discussed in detail the findings and their systematization during the analysis process, with the purpose to reduce the risk of misinterpretations due to the researchers' preunderstanding.
The analysing strategy consists of six steps (Graneheim & Lundman, 2004). The analysis started with multiple readings of the transcripts (a) before searching for content that described the participants' experience with using these tools and how they regarded the tools (b) followed by identification of meaning units (c). Then, the meaning units were condensed and coded (d) and subsequently interpreted (e) and sorted into preliminary categories and themes (f), which are threads of meaning running through the previous steps (Graneheim & Lundman, 2004). The steps up until categorization are description of "the content on a manifest level with low degree of interpretation," and from "theme" onwards, the process describes content on a latent level with a high degree of interpretation and abstraction (Graneheim, Lindgren, & Lundman, 2017). These steps do not suggest a linear analysis process; however, moving back and forth in the analysis were necessary to concentrate findings responding to our research question. Analysis resulted in three themes: (a) factual knowledge, (b) common languageincongruent understanding, and (c) gained confidencereduced expertise. These will be elaborated in the findings section.
This study focuses on a limited selection of standardized tools.
Including other tools would possibly have supplied additional insights to the study. However, narrowing the focus on specific tools may provide more concrete responses than talking generally about standardized tools. In this study, the CWS agencies differ in size and the length of their experience using the selected tools. This can give a range of insights.

| FINDINGS
Findings show that CWS professionals experienced that KF and COS-P strengthened their professional practice, but they also felt there were challenges. The three themes identified were present in relation to both tools, but to a different degree. These will be presented in the following.

| Factual knowledge
Participants reported that the tools to some degree helped them acquire new knowledge. This was particularly relevant to risk assessment and understanding the family situation in the light of theoretical knowledge, mainly psychological theories. They felt that a synthesis between theories and the observed family situation resulted in more knowledge, perceived as factual, about the family situation: We get more proof and richer description of the children … but also the parents-more about their skills in relation to maltreatment, such as physical abuse, substance abuse, mental health and mentalisation ability. We now have more questions for the parents, how they perceive and describe their child. … We learned to consider rationally why we were concerned. (R7) The CWS workers valued factors they perceived to be factual in their KF assessments. Such factual knowledge entailed a perception of tools providing answers to challenging questions, as one expressed, "COS-P provides me the answer of what good parenting is" (R37). In general, participants expressed that they relied on knowledge gained using the tools because they felt they had more evidence on which to base their understanding of the family.
Psychological knowledge, focusing on risk and protective factors, seem to be preferred as a knowledge base. Despite some being sceptical about the increased risk orientation, the general view was that tools were supportive to their work and helped them not only to describe the family situation, but also to link descriptions and observations with risk factors. They reported that focus on risk was gaining ground among the workers as well as managers at their workplace and that other professionals valued risk-oriented reports. Participants reported, for example, that other professionals took their work more seriously when they worked in accordance with the tools. This was especially evident when presenting a case before one of the Boards, which decide upon compulsory measures: The attorney picked up the Kvello book during her procedure; she raised it above her head, stating, 'This is not a coincidence!' It was so symbolic! Showing this is not just something only CWS believes, or a subjective opinion from a caseworker. It shows professional substance … She stated there are 10 risk factors present, and no protective factors! The opponent's attorney did not say a word … this unified us. (R19) This shows that risk factors are perceived as factual information about the child's situation. In addition, caseworkers trusted that their own judgement becomes transparent, and, consequently, others will agree so they gain credibility with other professionals. Participants experienced that the Board easily suggested COS-P as a solution in complex cases. However, participants were concerned about this because COS-P is not suited for multiple problems. These findings show that standardized tools are gaining ground and that CWS work using such tools is seen as more reliable by caseworkers and others.
Participants also presented limitations of both tools. They questioned whether the tools led to a biased understanding of parenting. Several found it challenging to use Western-informed tools in non-Western families because of the lack of cultural tailoring. For example, the parent interviews in both tools were not seen as appropriate when working with some migrant families: I believe it is related to culture … they have another perception of parenting and care … I am afraid we have different perceptions, which makes it very challenging to justify what we are doing. (R34) They also addressed the risk of misinterpreting a parent's answers because of language barriers. Consequently, participants found themselves in a dilemma when they knew these parents were likely to respond poorly because of their culture or language, and therefore, the process risked doing them injustice. Some social workers showed an awareness that when such tools are perceived as providing "factual" knowledge about the family, they can lead to biased understandings of the parents and the family situation.

| Common languageincongruent understanding
Another key theme was how both tools provided the workers with a common and professional language, consisting of new concepts and consistent descriptions of family situations. Participants highlighted how new concepts (e.g., risk factors and mentalization) and the tools' terminology helped them describe the child's situation better. For COS-P, this also included metaphors and symbols.
The "new" language aided communication with parents and other professionals. Furthermore, the common language deriving from the tools provided more consistency among team members and facilitated better descriptions when presenting a case to other professionals: [When] they ask about attachment, I can use the illustration of hands on the circle … And, if the mother uses substances, there are no hands available for the child, which enables us to picture the situation. We now have the skills and language to describe this. (R38) Despite enhanced common language, some participants experienced a lack of mutual understanding of the various concepts. This particularly concerned ambiguous risk factors. One participant stated, "We interpret concepts differently, for instance, what is substance abuse?" (R2).
They also reported uncertainty in how to handle conceptual discrepancies between the workers: I completed an assessment, the third in this case.
The caseworkers who conducted the two previous assessments put down different risk and protective factors, and there was no new information! (…) I refused to put these in the report, we could not present additional new factors, which could contribute to mistrust [from parents]. (R14) For families in contact with different CWS workers, incongruent assessments can be very confusing and upsetting, which in turn may affect the families' relationship to the CWS. Another issue was the rigidity of risk factors that they were not allowed to rephrase. This put the workers in a difficult position: I cannot include the word sexual abuse in their report if that's not a topic [during the assessment]. It is absurd and abusive to those receiving the report.

(R14)
Predefined phrasings and concepts may not always fit reality, and although they were not supposed to, a few participants did rephrase the risk factors when they felt it necessary. These findings suggest that when a perceived common language is taken for granted, and concepts and risk factors lack operationalisation, different interpretations of the same situation are likely to occur. The challenges in relation to language are most prominent for KF. Except for challenges in relation to migrant families, COS-P was perceived positively in terms of language.

| Gained confidencereduced expertise
Finally, participants experienced that both tools increased their professional confidence through a more focused practice and an increased theoretical proficiency particularly in relation to complex cases. Both tools helped them address difficult issues in conversations with the families: My job entails asking uncomfortable and sensitive personal questions … since the questions come from a guideline, it makes me feel more confident about asking these questions and more assured that I have the necessary information for assessing the risk to the child. This makes me feel more professional. (R21) This illustrates how participants trusted the guidelines and their ability to yield richer descriptions and thus "better" results. Consequently, they felt more professional.
Professional confidence also included acquiring a more structured and focused practice with less distraction from complexities. Hence, both tools facilitated better understandings of the family situation.
Additionally, KF provided better categorization of the information obtained: The information becomes so systematic. You begin at one end and go through it. It covers areas we have forgotten; it makes it much more transparent. You get so much more information … we ask more, see more and do more observations, which are valuable for assessing the situation. (R20) Some participants nuanced this by suggesting that more focus on some aspects diminishes focus on others: If I apply COS-P, there is no room to discuss the other matters they [the parents] experience … at the same time, it is problematic to limit what they can talk about, because it might be topics that are important for the parents. … If I am going to complete COS-P, I have to control the session. … I think that is very hard, because these are their concerns. (R31) CWS workers experienced that the tools challenged their professional ethos in which relationship and client perspective is important.
Despite the fact that both tools provide CWS professionals with proficiency and structure, which boost their professional confidence, some workers addressed how strict adherence to the guidelines also raised challenges. For example, several acknowledged the risk of making KF assessments too extensive resulting in an overload of information.
Consequently, the workload increased and contributed to exceeding the time limit of the assessment. Some were also concerned about parents disclosing more information than necessary and jeopardizing a relationship with parents based on trust. Participants described struggles with analysing the amount of information and did not feel they found sufficient guidance in the tools for dealing with this. Consequently, some felt the tools undermined their professional expertise: We were waiting for guidance on how to assess the information obtained… and maybe we used the framework instead of our own knowledge. In retrospect, we were actually fully capable of summarizing this adequately on our own. However, we became so set on the system, and in what order to do things, that we became incapable of acting. (R20) Overall, the findings suggest that CWS workers in these two offices have a strong reliance on the tools and view them as beneficial for their professional work in terms of applying theoretical knowledge in practice, acquiring a common language, and enhancing their professional confidence. Dilemmas and challenges were also addressed by the professionals, and they stated the importance of not applying the tools blindly, but instead relying on their own professional judgement.
Others emphasized that "one size does not fit all." When workers are confident in their role, they may more easily allow themselves to modify the tools in accordance with their professional ethos.

| DISCUSSION
The analysis shows that standardized tools (KF and COS-P) influence professional roles by guiding professional practice. The professionals generally felt that the tools contributed to enhanced professional competence and confidence. However, some did at times experience the tools as challenging their professional ethos and, as such, their professional role.

| Favouring explicit theoretical knowledge
CWS workers and other professionals recognized the tools as providing concrete and factual information about the families. Standardization, as such, becomes a procedure to reduce biases and contribute to objectivity. This represents a shift in social work towards a rulefollowing approach underpinned by theoretical knowledge and objectivity (Munro & Hardie, 2018). Such explicit theoretical codified knowledge (Polanyi & Sen, 2009[1961) appears to be the new normative standard for public professionalism at the cost of critical reflective practical knowledge that encompasses knowledge and skills that can be both explicit and tacit (Freidson, 2001).
Findings suggest that tools have the potential to make tacit knowledge explicit (Polanyi & Sen, 2009[1961) through a common psychological language offered by the tools. The participants experienced this as a support in their work, which increased their professional confidence. However, they had difficulties articulating the rationale and their expertise, known as the practical knowledge, beyond the codified knowledge deriving from the tools. As such, the tools do not seem to bridge the tacit and explicit dimension of knowledge. Freidson (2001) argues that knowledge in professional work is "embodied," which entails that the professionals must have an understanding of the rationalities of the knowledge constituted in the standards, as well as the skill to carry out the task. In this sense, is it not enough to "simply" learn the procedure of the standard, which some participants in this study experienced as problematic (e.g., "we do not know what to do with the obtained information"). Consequently, the professionals need to take into account the clients' needs, knowledge over "the particular case," as an attempt to avoid committing the client injustice. Accordingly, all forms of knowledge are considered relevant in a way that guides practice. However, as demonstrated in this study, the concepts and procedures deriving from the tools seemed to be the preferred forms of knowledge, without the professionals necessarily having internalized the theoretical knowledge situated in the tools. Findings suggest a risk of prioritizing explicit psychological knowledge embedded in the tools, above critical reflective and professional judgements and skills, the practical knowledge.
Moreover, findings suggest that professionals experienced that they could make "objective judgement" when applying the tools.
However, in line with Molander's (2016, p. 7) arguments, there is an intrinsic problem when a judgement entails a claim distinguishing true from false. First, there is an underlying assumption that knowledge is impersonal and value-free (Munro & Hardie, 2018). Yet, the guidelines favour specific psychological theories and a Western understanding of parenting. This particularly caused dilemmas for caseworkers working with minority families. Second, if a conclusion is perceived as true, the process leading to the conclusion also needs to be value-free and clearly explained. In this study, the guidelines were found difficult to operationalize (e.g., risk and protective factors and parent interview) and thus were open to multiple and value-laden interpretations. The guidelines failed to handle complexity and contextual variations, with a risk of overlooking the individual needs of families. Such simplifications along with professional equating judgements as truths may affect the relationship with the family negatively. However, professionals do experience an increased legitimacy from other professionals, endorsed by the use of explicit theoretical concepts (e.g., risk factors). An important question is whether factors that contribute to increased legitimacy from other professionals have the reverse effect on families in contact with CWS. Nevertheless, we argue that a narrowed knowledge base for CWS practice is problematic due to the complexity and contextual matters in CWS cases. Consequently, and in line with previous findings (Gillingham, 2011;Ponnert & Svensson, 2016;White et al., 2008), standardized tools run a risk of weakening the professional role, rather than strengthening it.

| Standardized procedures and professional accountability
Findings from this study aligns with previous research (e.g., Bartelink et al., 2015;Vis et al., 2016) showing that professionals' experience enhanced skills in yielding systematized thick descriptions through the tools. This is of importance for identifying children at risk. However, when CWS professionals follow the procedures "to the letter," they also run the risk of making assessments too extensive. At the same time, tools were found to restrict which topics were addressed, with a risk of disregarding contextual matters and themes that family members considered important.
As Thompson (2016) argues, professional accountability depends on critical reflective practice. He asserts a shift from reflective practice to a "rule-following" approach, which undermine professional expertise and challenge professional accountability. Instead, CWS professionals are inclined to adapt to a procedural accountability (Banks, 2009). Being accountable denotes the ability to account for decisions in a way that justifies actions (Banks, 2009;Molander, 2016). Our findings suggest a possible tension between procedural accountability and professional accountability. Despite relying on the tools were seemingly prominent among the professionals, some would overrule the procedural "protocol" when the standards conflicted with their professional ethos. Although standardized tools may strengthen the ability to describe a situation based on a theoretical vocabulary, the professionals do not necessarily strengthen a critical reflective practice in relation to how family situations are understood as well as in their application of knowledge.

| CONCLUSION
The professionals experienced that standardized tools strengthened their professional role through an explicit common language among professionals within CWS and in communication with collaborative partners. This led to increased legitimacy and enhanced their professional confidence. On the downside, the tools also led to frustrations in terms of increased workload and being caught in predefined actions, which challenged professional expertise. In conclusion, how standardized tools influence the professional role depends on the professional's ability to exercise critical reflection and professional judgement. As such, the tools become mechanisms of jurisdiction and legitimacy. However, when tools lead to a more procedural-focused approach together with a narrowed knowledge base, there is a risk of undermining professional expertise and doing families, as service receivers, injustice. Accordingly, the professional role, professional accountability, and client legitimacy may be weakened. To avoid this, we argue that paying attention to critical reflection and a broadened knowledge base are crucial for more productive actions for CWS practice.