Professionals’ tinkering with standardised tools: dynamics involving actors and tools in child welfare practices

ABSTRACT

This article aims to examine how child welfare professionals adapt standardised tools into practice. It focuses on how the professionals apply two commonly used standard tools in Norway and how they make them fit their daily practice. The research question is: How do professionals adapt standardised tools into the practices of the child welfare service? This is explored through the concepts of institutional work. The data stem from observation, client documents and semi-structured interviews with frontline professionals and managers in two child welfare offices. The data were analysed using thematic analysis. The findings show that new rules for practice and knowledge emerged when the tools were used. Moreover, the professionals modified them to suit their professional ethics and workload; here, the managers encouraged them to exercise discretion. We argue that how the tools were adapted depended on the institutional work of the child welfare professionals, through creation, maintenance and disruption. Consequently, there is a dynamic process between the actors and the tools, in which they both impinge on one another. Furthermore, frontline professionals still maintain substantial discretion and tinker with tools.

ABSTRAKT

Denne artikkelen har til hensikt å utforske hvordan de profesjonelle i barnevernet tilpasser standardiserte verktøy til sin praksis. Oppmerksomheten er rettet mot hvordan de anvender to standardiserte verktøy som er mye brukt i barnevernet i Norge, og hvordan de modifiserer verktøyene i sin daglige praksis. Forskningsspørsmålet som utforskes er dermed: Hvordan tilpasser de profesjonelle standardiserte verktøy til barnevernets praksis? Dette utforskes gjennom konseptene institusjonelt arbeid. Datamaterialet er innhentet gjennom observasjon, klient-dokumenter og semi-strukturerte intervjuer med ansatte i førstelinjetjenesten, og deres ledere, i to barneverntjenester. Datamaterialet ble analysert ved bruk av tematisk analyse. Studien viser at nye regler og ny kunnskap oppstår når verktøyene tas i bruk, og at de profesjonelle modifiserer dem for å tilpasse sin yrkesetikk og arbeidshverdag; herunder oppfordrer lederne de ansatte til å anvende skjønn. Vi argumenterer for at det institusjonelle arbeidet utført av profesjonelle barnevernsarbeidere gjennom skapelse, opprettholdelse og forstyrrelser påvirker hvordan verktøyene modifiseres. I så måte er det en dynamisk prosess mellom de profesjonelle og verktøyene som gjensidig påvirker hverandre. Studien viser at de profesjonelle fortsatt har stort rom for skjønnsutøvelse og at de modifiserer de standardiserte verktøyene.

KEYWORDS:

• Child welfare
• standardisation
• institutional work
• discretion

NØKKELORD:

• Barnevern
• standardisering
• institusjonelt arbeid
• skjønn

Introduction

In recent years, the child welfare service (CWS) in Norway, as in many other Western countries, has started using various standardised tools and frameworks (Christiansen et al., 2019; Møller et al., 2019; Vis et al., 2019). The purpose is to guide CWS professionals’ decision-making processes and interventions (Bartelink et al., 2015; Gillingham et al., 2017; Wike et al., 2014). Previous research on standardised tools in the CWS has been examined with various conclusions as to whether the tools are fit for the purpose (Sørensen, 2017; Vis et al., 2019; Wike et al., 2014). The use of assessment tools does not ensure uniform practice, either in relation to the information professionals use as a basis for assessments or for the prediction of risk (Gillingham et al., 2017; Sørensen, 2017; Vis et al., 2019). However, policy makers seem to be under the assumption that professionals use the tools as intended (Gillingham & Humphreys, 2009). The Norwegian government still recommends the use of standard tools in the CWS (Vis et al., 2019). This article aims to explore how standard tools influence existing CWS professionals’ practice.

Standardised tools have been heavily criticised for challenging professional practice (e.g. Gillingham & Humphreys, 2009; Munro, 2011). They provide guidelines with predetermined courses of action to regulate professionals’ decisions in order to enhance practice through scientific methods (Møller et al., 2019). In addition, they ensure effective, transparent and accountable services (Timmermans & Epstein, 2010). They can be understood as ‘procedural standards’ that determine the actions of professionals (Timmermans & Berg, 2010, p. 26) and are related to norms (Brunsson & Jacobsson, 2000).

Studies suggest that professionals are not passive receivers of standard rules; they may critically reflect on, modify or even reject the guidelines of the tools (Breit et al., 2016; Skillmark & Denvall, 2018). The argument is that standard tools do not capture every aspect of practice, which makes professionals tinker with the standards in different ways (Timmermans & Epstein, 2010). Some studies have shown that professionals sometimes ignore the tools or use them in conjunction with their own professional discretion (Gillingham & Humphreys, 2009; Sletten & Ellingsen, 2020). Discretion, which is a core value in professional work, is explained as an area of power where professionals have the authority to choose between lawful alternative actions based on their judgement (Molander et al., 2012). Several social work studies have demonstrated that standard tools challenge professionals’ discretionary space (e.g. Healy & Meagher, 2004; Ponnert & Svensson, 2016; Wike et al., 2014). It is argued that the tools prevent the development of professional expertise through compliance with management rather than meeting service user needs (Gillingham & Humphreys, 2009).

However, research has also shown that standardised tools are modifiable, thus enabling discretion (Evans, 2010; Gillingham et al., 2017; Høybye-Mortensen, 2013). Furthermore, professionals sometimes overrule the guidelines by replacing them with their professional expertise (Sletten & Ellingsen, 2020). Accordingly, the literature is ambiguous as to what extent more rule-bound practice (e.g. standardised tools) has curtailed professional discretion (Broadhurst et al., 2009; Djupvik et al., 2019; Evans, 2012). As a way to achieve their objectives professionals may apply discretion to negotiate formal rules (Ellis, 2011). With few exceptions (Breit et al., 2016; Gillingham et al., 2017), there is need for more research on how CWS professionals respond to tools and the relationship between tools and professional discretion (Evans, 2010). Changes do not solely depend on procedures and rules introduced by management, but also local institutional work (Breit et al., 2016).

We argue that the concept of institutional work (Lawrence & Suddaby, 2006) helps to explain and provide understanding of changes in professional practice that emerge from the use of standardised tools. Moreover, the concept of institutional work will enable elaboration on possible tensions between standardised tools and existing practice (Cloutier et al., 2015; Lawrence & Suddaby, 2006). This includes how the actors respond to these tensions through their actions and interactions, processes in which professional discretion is essential.

This study contributes to the ongoing debate about the interplay between standard tools and professional practice, i.e. how actors respond to tools and new modes of practice. The question pursued in this article is: How do professionals adapt standardised tools into the practice of the child welfare service? This implies a focus on the professionals’ actions and the rules that guide their practice, including the significance of professional discretion.

Context of the study

The Norwegian child welfare system is service-oriented with a child-centred approach (for further reading, see Skivenes & Søvig, 2017). This implies that voluntary or compulsory measures are needs-based to promote a healthy childhood, involving e.g. parental counselling, and not confined to maltreated children. The CWS is regulated through the Child Welfare Act (1992), and has incorporated the Convention of the Rights of the Child (CRC, 1989).

The Norwegian CWS has adopted a variety of standardised tools. This study investigates two standardised tools in the Norwegian CWS. The Kvello Assessment Framework (KF) and the Circle of Security (COS) are both widely used in the Norwegian CWS (Christiansen et al., 2019; Vis et al., 2019). Both tools provide a method of practice for frontline professionals to perform their daily work. They consist of guidelines with predetermined key tasks linked to scientific evidence, but the tools do not strictly qualify as evidence-based programmes. In this sense, the expert knowledge is rooted in the procedures constituting the standards (Brunsson & Jacobsson, 2000). The Norwegian government does not prescribe either of the tools as part of the CWS. They do however, advise other public family services to offer COS-P.

KF is a non-licensed decision-making assessment framework (Kvello, 2015). It shares some similarities with the Danish ICS (Integrated Children’s System) and the Swedish BBIC (Children’s needs in the centre) models, both originating from the British Assessment Framework (AF) (Christiansen et al., 2019). KF aims to guide CWS professionals in assessing the family situation with a variety of approaches, such as a risk assessment and structured parent interviews, which are key elements of KF. The KF framework promotes the division of information into various boxes, such as the child’s self-report and the caregiver’s understanding of the child. KF is not manual-based and is used in various forms by over 50% of local CWSs in Norway (Vis et al., 2019).

COS is a manual-based and licensed parental intervention programme originating from the USA. COS is rooted in attachment theory, aiming to enhance caregivers’ ability to provide emotional support to their child (Powell et al., 2013). The COS protocol consists of a manual, DVD, graphics and various metaphorical key concepts (Powell et al., 2013). There are different versions of the COS; this study is based on the eight-session parenting programme, COS-Parenting (COS-P). All COS-P therapists must attend a four-day training course to obtain certification. COS and COS-P are implemented across family services in Norway (Christiansen et al., 2019).

KF and COS-P are used in different phases of a CWS case and are thus complementary. Our aim is not to compare the tools. However, they both constitute examples of standardisation, given that our interest is in how CWS professionals use and regard standardised tools.

The concept of institutional work

Institutional work has gained importance in studying institutional change (Breit et al., 2016). This implies a practice perspective focusing on sets of practices in which professionals purposively engage independently and collectively. However, this does not imply that actors are free to act independently (Lawrence & Suddaby, 2006). When standardised tools are put into effect in CWS practice, the tools contribute to the creation, maintenance and disruption of institutions (Lawrence & Suddaby, 2006). Furthermore, through attention to institutional resistance, institutional work is viewed as actual ‘work’ that engages in challenging and negotiating current practices, beliefs and rules that may be in opposition to it (Cloutier et al., 2015; Lawrence et al., 2011). Institutional change is thus seen as an interplay between professionals who influence institutions through their discretionary space (Lawrence & Suddaby, 2006), and the organisation that aims to control and regulate the professionals (Cooper & Robson, 2006). Consequently, professionals are seen as both a mechanism for change and the main target for institutional change (Lawrence et al., 2011).

Lawrence and Suddaby (2006) differentiate between three broad categories of institutional work: creating, maintaining and disrupting institutions. Creating institutions entails actions in which frontline professionals and managers engage in reconstructing rules or changing categorisations within the meaning system. The latter includes theorising through the naming of new concepts. The adoption of a rule-following approach underpinned by objective knowledge and norms in traditional social work is one such example (Sletten & Ellingsen, 2020).

Maintaining institutions includes supporting, repairing or recreating social mechanisms to ensure that the existing practice remains. In this regard, actors strive to maintain systems or beliefs that seem to favour them (e.g. user involvement), which they do through their socially prescribed role (Lawrence et al., 2011). Professionals may, for example, combine their professional expertise with standard tools (Gillingham et al., 2017).

Disrupting institutions involves ‘attacking or undermining mechanisms that lead members to comply with institutions’ (Lawrence & Suddaby, 2006, p. 235). For instance, whether to use evidence-based practice or not is made into a question of professional ethics (Gibbs & Gambrill, 2002) and thereby undermines existing practice.

Method

We investigated the adaptation of two standardised tools, KF and COS-P. A qualitative case study design (Yin, 2014) was chosen with the aim of examining how professionals adapt these tools into practice. Our case context is that of the interplay between professionals and practice in local CWSs. To increase the possibility to identify patterns, the data collection was undertaken in two child welfare offices located in different regions of Norway (Braun & Clarke, 2006). ‘Office A’ had used the tools for about ten years, while ‘Office B’ had recently started to use them. Additionally, the two offices were organised differently; the work of Office A was more specialised. This was therefore an exemplifying case (Bryman, 2016), in which we analysed sets of practices in which the professionals engaged.

Data collection

The study drew on several data sources: fieldwork, interviews and documents. The document data were used in conjunction with observation and interview data to reveal possible misunderstandings during the analysis process (Bryman, 2016, p. 386). Collecting data over time, while combining various approaches, enabled a deeper understanding of the CWS professionals’ practice in using the tools.

The fieldwork comprised 51 days of observation in the offices over almost a year (April 2017 to March 2018). Observations were conducted during participation in day-to-day activities, meetings (e.g. internal meetings, group supervision, and client meetings), and informal talks with the CWS professionals. The observations were recorded through field notes written up the same day, which allowed for reflection and sampling of emerging topics for further focus. The researchers aimed to gain access to ‘backstage’ activities and to capture how the frontline professionals and managers applied the tools in real time in their daily practice, through both observations and unstructured discussions with the participants (Bryman, 2016). Central topics in the observations were descriptions of which professionals used which tools, when and how. The first author conducted the observations and to gain greater insight into the professionals’ practice she alternated between key informants and ‘regular’ informants (Bryman, 2016). The observational data served as a valuable foundation of background knowledge for use in the interviews to nuance the participants’ descriptions.

Interview data were elicited from a variety of qualitative interview approaches: (i) 25 semi-structured individual interviews (Office A: n = 16 and Office B: n = 3), (ii) eight semi-structured group interviews with two to five participants (Office A: n = 21 and Office B: n = 2), and (iii) several unstructured interviews in both offices (Bryman, 2016). The management teams allowed us access to these offices and recommended the most suitable team in the office for our study. One selection criterion was that participants used one of the tools in their daily practice. A total of 49 CWS staff consented to participate in the study (KF (n = 27), COS-P (n = 11) and management positions (n = 11)). All participants except one had a bachelor’s degree in social work/child welfare and some had additional education. Work experience varied from 1 year to over 30 years, which provided a variety of experiences in the use of the tools.

All semi-structured interviews lasted for 60–90 minutes and were conducted at the participants’ workplace during working hours. Both frontline professionals and managers were interviewed, and some participants were interviewed more than once. Of the 33 semi-structured interviews, 25 (individual and group) were recorded and transcribed verbatim, while 8 (individual) were documented through extensive note taking. Topics addressed included the participants’ understanding of standardisation, experience of using the tools, reasons why they used them, and whether and how they found the tools supportive for their work. The individual interviews allowed the participants to express their experience in depth without interference from others. The purpose of the group interviews was to encourage discussions on the same topics and to elaborate on themes emerging from the observations and individual interviews. Group interviews can discourage participants from expressing their opinion due to group pressure. To minimise power dynamics within groups, all groups consisted of participants working within the same team and holding the same position (Bryman, 2016).

Furthermore, we included 15 case investigation reports based on the KF framework, written by the CWS professionals. These documents were essential to understand how the professionals applied the tool. The investigation reports, which were all completed in May 2017, were randomly selected from different teams from the same office. The purpose was to see how they entered the information into the boxes of the framework. We were particularly interested in how they reported risk and protective factors and the structured parent interview.

Data analysis

The various data sources generated thick descriptions, which were analysed using thematic analysis (Braun & Clarke, 2006) supported by NVivo 11. To reduce the risk of misinterpretations due to the researchers’ preunderstandings, the authors discussed the categorisations and systematisation of the empirical data in detail during the analysis process. We looked for patterns of common meanings (Braun & Clarke, 2006) in relation to the actors’ behaviour and activities revealed through observations and in documents, and discussed this with the participants in interviews. Moreover, we identified actors who played a key role in either creating, maintaining or disrupting an institution. First, the first author read through the dataset and conducted initial coding in relation to the concept of institutional work, i.e. a deductive approach with strong linkage to theory (Yin, 2014). Second, we took a more inductive data-driven approach (Bryman, 2016) and developed codes and categories through multiple readings of the dataset. Accordingly, we do not totally subscribe to a deductive approach. This resulted in over 20 categories. Third, we reviewed the categories and made necessary changes and modifications. Finally, we refined the categories, which resulted in three themes: reshaping rules, changing knowledge and skills, and persisting with previous professional practices.

Ethics

This study was approved by the Norwegian Centre for Research Data (NSD, project number 53005, dated 16 March 2017). All participants were informed about the study and signed a written consent. Additionally, parents who were present at the client observations were all informed about the study and all gave consent. For the retrieved documents, we applied for special approval and permission was given by the Norwegian Directorate for Children, Youth and Family Affairs. These documents are highly sensitive and due to ethical challenges the number of case files was restricted. Only one of the researchers had access to the files. The limited number of case files is considered a limitation of this part of the study.

Findings

Findings show that CWS professionals and managers generated new sets of practices when they used the standardised tools. At the same time, they modified the tools introduced to practice by creating new rules to enable the tools to match their professional beliefs and work reality. The three themes identified are presented below.

Reshaping rules and challenging discretion

Participants reported finding the tools supportive for their work, and the tools generated sets of new rules for their practice. We identified two main rules that guided the professionals in how to perform their work. First, they developed new focal points for areas to assess or address, e.g. risk and protective factors (RPFs) and emotional care. These focal points were recognised as distinct rules that many of the participants followed. However, in office A, use of RPFs was mandatory; here, the management more clearly instructed caseworkers to use RPFs. The documents revealed that most of the reports included RPFs even though the RPFs were presented in different ways. In some reports, the link between the description of the family situation and the selected factors was made explicit, while in others, such information was lacking. There was also variation in whether and how emotional care and RPFs were assessed. This indicates some level of interpretation and exercise of discretion regarding the rule. The development of new focal points was also identified in the observations as illustrated in this field note from an investigation:

While caseworker R23 reads the report of concern, which contains much information, caseworker R20 takes notes. When R23 has finished reading, R20 states: ‘I wrote down all the risk factors’  […]  R20 does not have the list of risk factors in front of her; even so, she seems to have a good overview of the RPFs. (R20 and R23)

The participants also reported about other professionals who requested the use of these focal points:

The lawyers now ask specifically about the risk and protective factors in our report.  […] Another change is that we are now able to describe the [parents’] approach  […] by using mentalisation questions, but not the entire interview. I use mentalisation questions in almost every case.  […] we have integrated mentalisation in our way of thinking. (R2)

In this sense, new focal points were present in various situations and urged by various actors, and thus became important rules of the professionals’ practice.

Second, new rules of how the professionals should carry out their work, involving a more systematised and categorised practice, were developed. This was related to information and documentation, such as what information to provide about the families, along with detailed requirements on how to document and report on the acquired information. Many of the participants endorsed these activities that arose from the tools. Several stated that they had developed better ways to conduct assessments and describe the child’s situation (e.g. through observation and questions in the guidelines). When documenting and reporting the gathered information, participants had to fit the information into a predefined structure:

We now split the information in the report according to the boxes instead of describing the entire case. (R5)

Participants reported that they found this helpful for how to conduct their work. However, the client documents revealed no clear pattern in the kinds of information the professionals entered into the various boxes. Sometimes the information they entered was across and beyond the topics of the boxes. For example, summaries from meetings with parents (date and content) were recorded in the same box, which encompassed information on various topics. In this sense, the professionals used their discretion to record the narrative of the story rather than splitting the information according to the template. In turn, this generated more descriptions of the family situation, which increased the documentation, a rule that the participants, including the management, found time-consuming and challenging to follow.

Examples of a more systematic practice were also found in COS-P. The manual provided guidance on which topics to address and at what point in counselling, as explained by one participant:

We take one chapter at a time as they are presented in the manual, but we also go back as they are related. (R41)

These findings can be understood as a development of new rules for practice through the use of these tools, which is seen in the development of new focal points and procedures for practice. In turn, the professionals underlined that this supported as well as challenged their professional discretion.

Tools guiding knowledge and skills

The naming of new concepts and practices, where the concepts became part of the professionals’ daily practice, implied that the professionals drew upon the knowledge emerging from the tools. This was particularly evident in COS-P where metaphors and other linguistic techniques (e.g. notions like shark music) are embedded in the manual. Participants stated that they integrated the concepts into their way of thinking. Moreover, RPFs and mentalisation were also examples of naming of concepts. Participants communicated these concepts in informal talks, in client meetings, group supervision, and client documents. The professionals had copies of the COS-P circle and the RPFs on their bulletin board in their office. According to one participant, they also had copies of the RPFs in the toilet when KF was first introduced. In this sense, these concepts represented new knowledge and helped to adapt these tools into their daily practice. In this way, the tools guided the professionals’ ways of seeing and acting that in turn may have influenced their discretionary activity. This is illustrated in a field note from an intake meeting:

A parent, caseworker (CW) and a supervisor (S) are present. After the parent has described his child, S talks about how to relate to the child emotionally, which is said to be important. S seems to be referring to COS-P when she asks, ‘Are you available for me now if the child says Daddy/Mummy’ but without mentioning COS-P explicitly. Then the parent starts talking about how he feels stuck in his parenting, to which S replies, ‘maybe you have some shark music as we say in COS-P’. The parent does not object to this. After the meeting, I talk to S about what she thought of this meeting. S replies that she recommends COS-P group or individual counselling, saying, ‘The parent has some shark music’. (R36)

Enhanced competency on attachment and assessment of the family situation were other skills the participants reported having acquired through these new concepts. Participants stated that the concepts directed their attention and helped them to stay focused on different matters, e.g. emotional care, and provided better descriptions of the family situations. This indicates that the tools influence the body of knowledge the professionals draw upon in their daily practice, and thus their professional work. They argued that the new competency enhanced their ability to assess the family situation as well as to communicate their concerns to parents and other professionals, as explained by one of the participants:

When I write the report … and start assessing, then it [the child’s situation] becomes more apparent. It made me realise that there were nine risk factors …, which made me more aware of how serious the situation is. (R5)

These findings, supported by documents, observations and other interviews, illustrate that the new concepts were perceived as supportive, providing new skills, and were on the CWS professionals’ agenda. However, participants also presented some limitations of the tools and mentioned their concerns about being too rigid at the expense of parents’ expressed needs, which challenged their professional ethics. Furthermore, there was insufficient clarity in the framework, and some participants questioned how to conduct their work:

I don’t know what to do with all these facts, this information that I’ve obtained. How should I summarise, what to look for and ultimately, how should I assess this? […] Are we even trained for this work? (R21)

Uncertainty as to how to analyse seemed to be a common challenge among the professionals working with KF, and was particular evident in relation to RPFs and mentalisation interviews. Participants, including managers, emphasised that some risk factors were ambiguous and thus challenging to operationalise, e.g. a bad neighbourhood. Moreover, they did not necessarily agree that all risk factors actually constituted a risk, e.g. divorced parents. This illustrates the tension between the tools and professional knowledge. Nevertheless, the findings do suggest that the tools, when found supportive, generated new practices through new knowledge and skills.

Shared commitment

The analysis also revealed that much of the professionals’ previous practice continued, advocated by both the professionals and the managers. Participants elaborated on how the KF made investigations more extensive than intended, and thus challenged the formal rules set by the Child Welfare Act. Some participants therefore disregarded the procedures, while others expressed a need to adjust the KF. This was in order to make their workload manageable, but also to enable investigations to comply with legislation. The management that shared the professionals’ concern adjusted the KF. In Office A, the management simplified the framework:

It’s a problem that we investigate areas not relevant to the case and we’re concerned about the long reports … We resolved this by initiating ‘speedy investigation’. So in those cases we have in a sense moved away from KF.  […] We’ve got better at deciding when a full investigation is necessary. (R18)

It was a common view among the professionals and the management that the KF was better suited for complex cases. Hence, the managers also undertook a discretionary role. The managers in office A developed a local manual with guidelines on which areas to investigate depending on the case. However, the documents revealed that the prescribed areas were sometimes not reported on in the documents. This shows that the professionals did not always follow the local guidelines. Considering that the managers approved the reports, this suggests that the management supported the professionals’ deviation from the local guidelines. The following quote illustrates how management supported such individual tailoring:

I stress that we have to make individual assessments in each case, of what areas to investigate. […] My team uses KF very differently. That is ok with me; they must be their own master, because we all work towards the same goal. (R16)

As for COS-P, the professionals reported two types of challenges that were closely linked. The first involved a discrepancy between what they experienced as clients’ needs and topics addressed in the COS-P guideline. Some participants reported allowing parents to address topics that were not part of the COS-P protocol. The second challenge was that they found that COS-P did not fit every family. Since COS-P lacks a set of intake criteria, the professionals, supported by the management, developed local intake norms based on their professional judgement. They expressed this view in group guidance sessions, informal talks, client meetings, and in the interviews:

Some families don’t benefit from COS-P. Parents that are cognitively weak, if they don’t have the ability to give feedback, or aren’t able to reflect. Then we must apply more practical approaches. (R31)

To deal with these challenges, participants found it necessary to deviate from the tools by combining their professional discretion with the tools.

The above findings show a need for making local adjustments of the tools to adhere to formal rules, the resources of the CWS and professional ethics. Moreover, the findings reveal that neither KF nor COS-P fits all clients or situations, which implies a need for individual tailoring of the tools. Rather than attempting to curtail the professionals’ discretion, the management supported and even encouraged them to modify the tools according to their discretion.

Discussion

The concept of institutional work enables us to explain and nuance how new local practices emerged as well as how many of the existing practices were maintained through professionals exercising discretion. Our main argument is that the professionals and the managers played an important role by using professional discretion in reshaping new practices, as well as using the tools in a dynamic manner.

Creating and disrupting practice – challenging professional discretion

We have identified two key forms of creation in relation to how the professionals and the managers responded to the tools. First, they created new rule structures for their practice. This implies that the professionals adapted the focal points of the tools, e.g. the RPFs and emotional care. The tools also influenced how they proceeded with their work. Accordingly, the tools guided their actions as to what information to pursue and how to record their work and in that way defined what was relevant to include. In this sense, the tools may be understood as a belt of restriction of the participants’ behaviour (Molander et al., 2012) and thus an attempt to limit their professional discretion. However, the findings suggest that the professionals found great support in the structure, where they experienced the tools as promoting consistency and quality in their work, and may therefore have been more willing to comply with the rules of the standard. In turn, this contributed to supporting the new institution, i.e. an institution with new rules and new concepts (Lawrence & Suddaby, 2006) and increasing the legitimacy of the professionals in relation to an external audience (Lawrence & Suddaby, 2006; Timmermans & Berg, 2010). The concepts of RPF and COS-P played an important role here. In this sense, the tools are not regarded as anti-professional (Robinson, 2003).

Moreover, the previously common narrative structure was to a certain extent disrupted as the professionals presented information according to the structure of the framework. However, the findings revealed that the information was not always presented in relation to the predetermined categories. This indicates that the professionals did to some degree protect the initial narrative structure and modified the framework in a way that they justified with reference to their work reality (White et al., 2008). This has also been found by other scholars who argue that rules need to be interpreted and negotiated into the local contexts (Ellis, 2011; Evans, 2010), which will leave room for the use of discretion (Molander et al., 2012).

Second, the professionals adapted new concepts and theoretical knowledge embedded in the tools, including new skills that informed their actions. The concepts and the new skills were perceived as meaningful and complemented their existing knowledge in a way that enhanced their competency (Bartelink et al., 2015; Robinson, 2003; Vis et al., 2019). According to Lawrence and Suddaby (2006), such complementary meaning systems increase the possibility to create new institutions, as the present study indicates. The new concepts and focal points (e.g. emotional care and RPF) were communicated in various settings, and when such concepts are repeatedly mentioned they became part of the daily practice (Lawrence & Suddaby, 2006). This provided a foundation for standard practice, and thus an increase in psychological knowledge emerging from the tools. Hence, the tools contributed to the diffusion of new norms and practices, which in turn may have influenced the discretionary activity through the professionals’ way of seeing the case at hand (Jobling, 2020). In this way, the use of the tools represented a shift towards constraining the professionals’ knowledge base (e.g. the use of RPFs), favouring a risk approach in addition to scientific psychological knowledge (Sletten & Ellingsen, 2020). In this regard, the findings suggest that the professionals’ discretion was restricted by the standardised tools (Molander et al., 2012). Based on the above discussion, we therefore argue that this is a case of creation in conjunction with disruption, as it is argued that creation is strongly linked to the disruption of institutions (Breit et al., 2016). Nevertheless, findings also show that even though practice is rule-based it is not necessarily rule-bound, as rules need to be interpreted into local contexts (Evans, 2010).

Maintaining practice through discretion

In line with previous research (Breit et al., 2016; Gillingham & Humphreys, 2009), our findings show that the professionals combined the KF and COS-P tools with their professional expertise, thus modifying the tools. This was particularly evident in relation to deciding in which cases the tools should be applied, but also when resolving professional dilemmas that emerged as a result of the clash between rules of the tools and the needs of the client. The professionals took a critical approach to the tools in defending their ethical practice (Jobling, 2020). The management mostly supported the modifications in an attempt to comply with professional ethics, and thus subscribed to a professional practice through a shared professional commitment (Evans, 2010). In this respect, the standard tools confronted established institutionalised structures (Cloutier et al., 2015), which was solved by the professionals modifying the tools through their discretionary power. In addition, the professionals took control of the guidelines as well as instructions from management, and altered the rules of the tools when they were found to be at odds with their professional ethics. The tools were thus used as an option rather than as a true standard (Breit et al., 2016; Timmermans & Berg, 2010). The professionals were therefore disinclined to move away from their professional expertise (Munro, 2011). In this sense, much of their discretionary practice was maintained, and the tools assumed professional knowledge (Jobling, 2020). This shows that ‘the presence of rules does not mean the absence of freedom’ (Evans, 2010, p. 62).

Moreover, the professionals demonstrated a particular tendency to modify the tools when they challenged legislation and when they led to a heavier workload. The former indicates that the tools confronted an existing institutional structure manifested by the Child Welfare Act (1992), which works as a coercive barrier to change (Lawrence et al., 2011). In the latter case, the professionals, including the management, deviated from the KF framework, as it became too extensive in relation to the objective, as supported by previous research (e.g. Vis et al., 2019). In addition, lack of resources and discrepancies with legislation led the professionals to develop different versions of investigations. As has been pointed out, there are multiple ways for professionals to respond to a policy, or in this case a standard (Jobling, 2020). In this way, standardised tools pulled the professionals in conflicting directions in their attempt to modify the tools. The professionals, supported by the management, solved this dilemma by using their discretion (Molander et al., 2012). This brought into play their own normative standards of how to work. In doing so, they tried to resolve the tools’ ambiguities, address their inaccuracies and give priority to certain components. This raises the question of whether the KF tool is fit for its purpose and in this sense the professionals tinkered with the tool (Timmermans & Epstein, 2010), by exercising their professional discretion. Accordingly, standard tools do not necessarily change practice, at least not in line with their original purpose (Brunsson & Jacobsson, 2000).

In conclusion, we argue that how professionals adapt standardised tools depends on local institutional work undertaken by frontline professionals and management. Our findings shed new light on the interplay between standardisation and discretion. Our study shows that new practices were created as long as professionals found the tools to be meaningful for their practice, which may in turn have increased their legitimacy and limited their discretionary activity. The tools may constrain their knowledge base and favour one kind of knowledge, here psychological knowledge, which would seem to be a negative development. On the other hand, we find that many existing practices were maintained when the tools were found to be at odds with professional ethics. In this respect, standardised tools do not necessarily restrict the discretion of the frontline professional. Here, there seems to be a shared commitment between professionals and managers to comply with professional ethics, especially when there is a risk of treating families inequitably. Consequently, we argue that there is a dynamic process between the tools and the actors, in which they impinge on one another. We also find that, although tools may tend to restrict their discretion, frontline professionals and management still maintain substantial discretion and tinker with tools.

Acknowledgements

We would like to thank our colleagues in our research group at Østfold University College for helpful comments and suggestions.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Marina S. Sletten

Marina S. Sletten is a Ph.D. candidate in Social Work at Øsfold University College, in association with University of Stavanger. Her research interests are on child welfare, discretion, professionalism, professional practice, standardisation, and interprofessional collaboration. Beside the Ph.D. project that focus on reconfiguration of professional practice in child welfare services, she is engaged in a project that aims to develop the competence of frontline professionals in local child welfare services.

Catharina Bjørkquist

Catharina Bjørkquist is Professor in political science at Østfold University College. Her research interests are on health and social policies, work, organisations and coordination of health and welfare services and hospital management. She is engaged in projects that focus on local innovation processes including digitalisation, development of health and welfare services and user involvement.