A dilemmatic space: A qualitative study about teachers’ agency working with early identification of special needs

ABSTRACT

The present study focuses on challenges to the work of teachers as a result of a Swedish educational reform that mandates schools guarantee pupils receive early learning support concerning reading, writing, and number sense. The outcome of reforms is connected to how teachers achieve agency in their work , and we use the ecological model of teacher agency as an overarching theoretical framework and analytical dimension. The study involved interviews with 14 teachers, and the focus was on a) how teachers identify pupils in need of support and how this support is carried out, (b) how teachers would like this identification to be carried out, and how teachers would like to support pupils with weak number sense. Teachers described a ‘dilemmatic space’ (cf. Fransson and Grannäs 2013), which affects how they achieved agency in developing pupils’ number sense. The dilemmatic space described by teachers in this study is framed by necessary priorities pertaining to the cultural, structural, and material conditions of the current situation. According to the study’s results, whether teachers should perform the assessment merits discussion. The study concludes there is a risk that conducting assessments and offering support risks cancelling each other out.

KEYWORDS:

• special education
• teacher agency
• dilemmatic space
• early identification of special needs

Introduction

The present study focuses on challenges in teachers’ work resulting from Swedish education reform. The reform is called the ‘early learning support guarantee’ (ELSG), which is the term used in this text. ELSG mandates schools guarantee pupils receive early learning support concerning reading, writing, and number sense. In this study, early identification and support for developing number sense is in focus. This is interesting from a special education perspective, which is defined here as any measures in teaching taken to compensate for differences among pupils in need of special support in learning specific content. This includes the fact that ‘the notion of regular teachers providing education to a diversity of pupils is relatively prevalent in Sweden, particularly as an organisational matter. Thus, teachers are intended to be the implementers of inclusion in practice’ (Miškolci, Magnússon and Nilholm, 2021, 562). The special education perspective includes identification of at-risk pupils and making adaptions to the teaching in the classroom, as well as intensified one-to-one teaching, if needed. Consequently, the focus here is on mandates placed on mainstream teachers to identify pupils with special needs early on and adapt their teaching to meet these demands. In practice, this is often done in collaboration with special education teachers (Takala et al. 2015). The reform was formulated in 2019 and includes a mandatory assignment for teachers in grade one in Sweden. The reform stipulates that teachers must identify, teach, and evaluate pupils’ number sense (Prop 2017/18:195), and focus on identifying pupils with special education needs. The reform is rooted in the need to improve Swedish pupils’ academic results in mathematics by providing adequate support beginning in first grade. Mathematics results have steadily been declining in Sweden, and the reform is designed to ameliorate this problem (Prop 2017, 18).

On the other hand, if the goals of the educational reform will be realised depends on teachers’ engagement and capacity to work in accordance with the reform’s purpose (Biesta, Priestley, and Robinson 2015; Priestley et al. 2015). Reforms involve increased testing of pupils’ knowledge, which in previous studies has been shown to limit teachers’ agency in their work (Leijen, Pedaste, and Lepp 2020; Mirra and Morrell 2011). In Sweden, programmes have been initiated by the state supporting teachers’ professional development (Kirsten and Wermke 2017). The ambition of these programmes was to increase pupils’ performance. However, teachers discussed their need to comply with the instructions, rather than make any meaningful adaptations (Kirsten 2020). In addition, Mirra and Morrell (2011) conclude that the fact that teachers must conduct standardised testing to measure pupils’ knowledge limits their professionalism and autonomy.

It is relevant to investigate how a reform influences teachers’ work, and the challenges initiated by said reform. In addition, if a reform is intended to have an influence on pupils’ results, it is interesting to investigate what impact that reform has on pupils’ ability to receive support. For pupils to succeed in mathematics education, number sense is a pivotal starting point. A pupil’s basic number sense at the start of school has been shown to be related to their future mathematics development (Andrews and Sayers 2015). Pupils’ basic number sense is thoroughly evaluated in the ELSG reform. This is thought to inform the implementation of teaching support adequate to meet pupils’ needs.

In the study, we have adopted the concept ‘teacher agency’, as it has been formulated and applied in educational research (Priestley, Minty, and Eager 2014; Toom, Pyhältö, and Rust 2015). ‘Teacher agency’ concerns how teachers achieve temporally constructed engagement within a structure to create possible courses of action to respond to dilemmas or problems they face while teaching (Biesta, Priestley, and Robinson 2015). The concept stems from a pragmatist framework regarding how people respond to problems they face and can, in addition, be used to resist external rules and requirements. Consequently, this applies when said requirements do not conform to what teachers consider to be professionally defensible actions (Dovemark 2010; Molla and Nolan 2020).

From a special education perspective, it is motivated to investigate teachers’ experiences identifying pupils’ needs, implementing adaptations, and acting in line with their ambitions to enact teaching that would enable learning for all pupils (Norwich 1993). To achieve inclusion in the classroom, special education deals with how teaching can be organised to better meet the variation of needs among pupils (Göransson, Nilholm, and Karlsson 2011). Teachers’ special educational engagement influences how support is designed in regular classrooms. Special education in classrooms is based on the identification of pupils’ needs. Hence, the variation of needs in the classroom might lead to professional dilemmas for teachers (Norwich 1993).

The aim of the study is to investigate teachers’ experiences working under the ELSG reform with a focus on early identification of pupils’ number sense. The practical implementation of the reform involves assessment work that will take place in parallel with the teacher’s regular teaching in grade one. Our study focuses on teachers’ ability to achieve agency in ELSG reform work. This is investigated through the following specific research questions: What challenges do teachers encounter in their work implementing ELSG reform? In what respects do teachers achieve agency in dealing with the professional challenges imposed on them by the ELSG reform?

Background

The Swedish Education Act (2010, 800) stipulates that all pupils must be given support at three levels to meet knowledge requirements. The three levels include: 1) guidance and stimulation, and if this is not enough for the pupils to meet the requirements extra adaptions (2) are also offered in the regular teaching setting. If pupils still do not meet the requirements, special support (3) should be provided (National Agency for Education 2022a). However, the Swedish School Inspectorate (2016) concluded in a quality review of schools’ work in extra adaptations and special support that this process starts too late. The ELSG was introduced to ensure that pupils are given early support measures, if necessary, for the development of basic language, reading, writing, and number sense.

The assessment of pupils’ number sense under the ELSG begins with a survey in preschool class and ends with conducting and following up national tests in grade three. In addition to these assessments, teachers in grade one must also assess pupils’ number sense by following the Swedish National Agency for Education’s assessment manual. The manual includes, among other things, a script for an assessment interview. During the interview, each pupil must orally answer 11 questions concerning number sense (Swedish National Agency for Education 2022b). Support for the pupil should then primarily be designed as extra adaptations (level 2). Therefore, said support is the responsibility of teachers working in regular education, and is not designed to primarily involve special education teachers. If a pupil shows a serious lack of number sense, they should promptly be given more comprehensive support, also referred to as ‘special support’ (level 3). This task is, in turn, assigned to the special education needs teacher (National Agency for Education 2022b). In general, special education teachers work directly with pupils in need of special support and consult teachers about their teaching. This includes special education teachers working with pupils in both segregated and inclusive settings (Takala et al. 2015), while mainstream teachers are supposed to teach in a way that offers guidance and stimulation to the full diversity of pupils in their classroom.

In 2021, the Swedish National Agency for Education conducted an evaluation of the work involved in the compulsory assessment in grade one. The results show that the assessment material is viewed with legitimacy among teachers. However, two problem areas were highlighted in the report: (1) teachers’ negative experiences trying to re-organise their teaching on their own based on the information that came from the assessment interview, and (2) a concern for the class situation during the time the teacher is absent (National Agency for Education 2021).

This study investigates challenges that teachers describe in their work on the mandatory assignment to assess pupils’ number sense at the start of school. In the next section, previous studies are presented based on aspects central to this study: (1) reform work in school, and (2) the professional dilemmas of teachers engaged in the identification of pupils’ support needs.

Teachers’ agency and professional decision-making in relation to reforms

Different types of policy documents create context for teachers’ tasks, but contextual factors are also crucial in implementing reforms (Ball, Maguire, and Braun 2011). The expectations placed on teachers have undergone major changes, not least through recent decades’ educational reforms and the pressure of higher goal achievement (Valli and Buese 2007). Teachers need to grapple with expectations of their professional practice, whether these demands are set by teachers or by policymakers (Tan, Chang, and Teng 2015). In addition, factors such as positive and trustworthy relationships with peers and members of the school leadership team constrain or enable teacher agency related to assessment (Poulton 2020). Tickle (2001) argues that teacher professionalism is defined according to what teachers themselves consider to be important in their work, based on their experiences in practice and their personal background. According to Toom, Pyhältö, and Rust (2015), teachers achieve agency when faced with dilemmas and uncertainties in their pedagogical practice.

Professional dilemmas of teachers engaged in the identification of pupils’ support needs

The fact that teachers are expected to carry out new reforms in parallel with their already-ongoing teaching creates pedagogical dilemmas (Jonasson, Mäkitalo, and Nielsen 2015). A dilemma, unlike a problem, is a problematic situation that does not have a solution without any negative effects (Flett and Wallace 2005). There are previous studies that show that new guidelines for teachers’ work imposed by reforms lead to limitations to a teacher’s professional work (Avis 2005). Fransson and Grannäs (2013) describe how dilemmas arise between the individual and the context. Furthermore, they argue that teachers are constantly in a ‘dilemmatic space’ that affects their decision-making. In their daily work, teachers face dilemmas in relation to expectations, competing goals, or questions about who should be responsible for pupils in need. Aspects that constitute a dilemmatic space are subjective. The conceptualisation of ‘dilemmatic space’ can be used to illustrate and deepen our understanding of teachers’ complex work situations (Fransson and Grannäs 2013). Teachers must manage the dilemma of using available resources (time, energy, and skills) in order to fulfil their administrative goals following available guidelines, laws, and policies, and meet pupils’ educational needs (Fransson and Grannäs 2013). The educational context is dynamic, complex, and unstable, which leads to changed conditions for teachers’ work (Day et al. 2006). How conditions might change is illustrated by the implementation of the ELSG reform.

Teachers need to achieve agency to implement reforms in daily pedagogical practice (Day, Harris, and Hadfield 2000). Consequently, teachers’ intention to deal with a dilemma can lead to additional dilemmas (Fransson and Grannäs 2013). This study aims to investigate teachers’ experiences working under ELSG reform and its stipulated focus on early identification of pupils’ number sense. This new assignment means that teachers, in order to fulfil the purpose of the reform, need to establish an organisation centred around identifying pupils with poor number sense and implementing special support for the pupils who need it.

Theoretical framework

As an analytical dimension, we use the ecological model of teacher agency, described by Priestley, Biesta, and Robinson (2016). Teaching places high demands on the teacher as an independent and active agent, both at school in general and in the classroom (Toom, Pyhältö, and Rust 2015). Teacher agency can be described as a teacher’s ability to act professionally and how they act in response to problematic situations that arise in a professional context (Biesta and Tedder 2007). This includes how teachers make different considerations in relation to their work and how these, in turn, affect their perceived ability to act. Priestley et al. (2012) provide examples of how teachers’ personal and professional experiences shape their agency. Teachers’ opportunities to act are influenced not only by their individual capacity, but also by contextual conditions in their work. A teacher may have good skills and knowledge and strive to provide their pupils with good education, but overly restrictive contextual conditions can keep them from reaching their goals. The conclusion that Priestley et al. (2012) draw is that reforms should focus on developing contextual conditions to enable teachers’ agency.

According to the ecological model of agency, three dimensions interact: (1) the iterative dimension, where the individual´s life experience and knowledge are accommodated; (2) the projective dimension—i.e. the individual´s future direction – and (3) the practical – evaluative dimension, which includes opportunities and limitations within the current situation. Achieving agency is, therefore, a result of reactivated experiences and goals, in interaction with the cultural, structural, and material conditions of the current situation (Biesta and Tedder 2007; Emirbayer and Mische 1998). This means that an individual does not ‘have’ agency, but agency is achieved in the transaction between the subjective and temporal aspects described above. In the interplay of cultural, structural, and material conditions of the current situation, an individual’s subjectivity involves having the ability to make principled choices in relation to a problem. We chose the concept of teacher agency as an overarching theoretical framework and analytical dimension, since we are interested in investigating teachers’ experiences working under ELSG reform, and how they need to establish an organisation centred around identifying pupils’ weak number sense and implementing special support for pupils who need it. Accordingly, teachers must achieve agency to fulfil the reform.

Method

A qualitative interview study was conducted due to our interest in teachers’ experiences and perspectives on identifying pupils’ needs under the present reform. Thematic analysis was utilised to analyse the data.

Participants

The study involved 14 teachers who, at the time of the interviews, were teaching mathematics in grade 1. All participants are qualified mathematics teachers. The participants’ experience working as teachers ranges from 1–25 years. At the time of the study, the participants were active in seven schools with between 85–500 pupils, in three different municipalities. At most, three participants worked at the same school.

Data collection

Data was collected through semi-structured interviews using an interview guide. The questions focused on: (a) how teachers identify pupils in need of support, (b) how teachers would like said identification to be carried out, c) how pupils who show weak number sense are supported, and d) how teachers would like to support pupils with weak number sense.

The interviews were conducted face-to-face at the participants’ workplace, or through an online video-conference service, at a time that suited the participants. The first author conducted all interviews and follow-up questions, returned to the claims made, and, if needed, asked for clarifications. These strategies were intended to result in a climate of open discussion and allow participants to freely develop their reasoning in the interview (Hiller and Diluzio 2004). The duration of the interviews ranged from 33 to 60.5 minutes (M = 40.5 min.). All interviews were transcribed verbatim, and pseudonyms were given to the participants to ensure confidentiality. Questions concerning how teachers gave pupils with weak number sense additional teaching were also a part of the interview. These questions were analysed and reported separately. This was done specifically to investigate how teachers described the adaptations they made to their teaching as a consequence of the assessment.

Data analyses

Thematic analysis is a useful method for analysing interview transcriptions (Braun and Clarke 2006). Therefore, once our data was collected and transcribed, thematic analysis was conducted where repetitions, patterns, differences, and similarities were coded and themes were constructed (Braun and Clarke 2006). As the starting point of the study, teachers’ choices and agency were the main focal points of the analysis. In accordance with Braun and Clarke (2006), the analysis was carried out in the following six phases:

1. transcripts of the interviews were read repeatedly to get to know the material.

2. The transcribed text was coded. Codes were created to summarise the meaning of a segment in the interview. Expressions that signalled different types of challenges, priorities, and dilemmas were highlighted. Examples of codes generated were: to teach or to assess, pupils must be assessed but also feel comfortable, and the best way to share information about the pupils’ needs.

3. The codes generated were then grouped together as themes. The theme ‘control’ (described below) consisted of codes such as: to get the bigger picture, who needs to know the children best, what is important information for teaching.

4. We reviewed the themes created in relation to the entire text and collected excerpts that exemplified each theme. Here, specific dilemmas emerged in teachers’ experiences.

5. Themes were defined and named, and

6. the text was compiled.

The first author took the lead in analysis. The second author reviewed the phases of the analysis and critically reviewed the process. Thematic analysis was chosen since it is a suitable method to find ‘repeated patterns of meaning’ (Braun and Clarke 2006, 86).

Ethical considerations

The study followed the Swedish Research Council’s research ethical principles (2017) in terms of giving consent to participation, usage of data, anonymity, and confidentiality. All teachers in the study were provided with information about the purpose of the study; first orally, and then in writing. All participants have been anonymised in terms of name and gender. Information about individual pupils has not been requested or investigated.

Findings

The main theme related to teachers’ work pertaining to the ELSG found in the study can be labelled: the right profession doing the right thing. This theme is further divided into three subthemes: control, relationships, and temporal validity of the assessment result. The current findings contribute to our understanding of how teachers achieve agency and face dilemmas during a reform change. In this study, the assessment interview is conducted either by a special needs teacher or a class teacher. The teachers expressed varying levels of satisfaction with their organisation. Regardless of how well they reported that the organisation was working, their answers highlighted dilemmas they had faced in carrying out the ELSG reform. In the text, special education needs teachers and teachers are described as different professions.

The right profession doing the right thing

The main theme was related to the dilemma of determining who is responsible for pupils in need of support. That included determining whether the class teacher or the special education needs teacher should conduct interviews to map pupils’ number sense. The analysis showed that this decision leads to dilemmas that were highlighted when teachers explained how they were simultaneously trying to be available and provide the support pupils required. Even if the profession doing the interview varied, there was a common goal in engaging in ELSG, which Åsa exemplifies in the following excerpt:

The whole point, as I see it… [is] someone who will then work with the pupils. Why should someone else assess and know everything about these pupils—it is much better that I get the information directly. It has to be someone who will be involved, more or less, in our work.

Åsa described that it was important that the person conducting the assessment is the same person that will be responsible for the follow-up teaching, so that they knew what type of support the pupils require. The dilemmas resulted were related to how teachers achieved agency, especially when it came to having control of the situation created by the ELSG reforms.

Control

When teachers described how the interviews were arranged at their school, the analysis illustrated that they, in different ways, aimed to be in control of information about pupils’ level of number sense. At the same time, teachers wanted to exercise control over the work conducted in the classroom. Lisa, who used to do the support assessment herself, now shares this task with the special needs teacher. Lisa reported she was missing out on forming her own opinion of her pupils’ level of knowledge:

I enjoyed carrying out the support assessment with all the children because I then felt that I had complete oversight over their number sense. I feel that it is very good for us [class teachers, authors’ note] to conduct the support assessment, as I then get a very clear picture of the pupils’ abilities.

She claimed that she had had better insight into pupils’ number sense in the past, whereas now she has to choose between focusing on the group or the individual. According to Lisa, neither option was optimal. Maria would like to perform the assessment on her own, but at the same time, she wanted to be responsible for the teaching in the classroom. The SEN teacher conducted the assessment with Maria’s pupils. Maria reported there could be a solution, but noted that this would also entail dilemmatic aspects:

As the class teacher, I would like to play a bigger role in this [assessing pupils], at the start of the first year. I believe it would give me so much/ … /If she [the special needs teacher, authors’ note] could have traded places with me and been present in the classroom during this time or something like that.

This arrangement was also present at Linda’s school, where SEN teachers carried out the assessments. This led to Linda not engaging with the assessment results, and she said that the results did not affect her teaching. If Linda could decide for herself, it would look different:

I get to conduct every single step, with each pupil. Because I want to be completely in charge, I’m a control freak. I have to say to myself ‘I can’t do that. The special needs teachers will have to do it’. So, I have to kind of switch to ‘she has to take over that now’. I can’t try and come in and get a full picture, I’m not able to do that. But I do want the full picture, not just pieces of it.

Linda was frustrated about not being able to form her own opinion of the pupils’ abilities. She described not wanting to leave her class to a substitute teacher. In this case, she could not get a full picture of the assessment results, since these were conducted by someone else.

The new reform is supposed to be included in the teachers’ already-ongoing practice, it is necessary for them to achieve agency to organise the implementation of the reform. Regardless of which organisational method they chose, the teacher told us that the assessment interview was not only about which profession (class teacher or SEN teacher) should be aware of the pupils’ number sense, but also for whom it was most important to have the opportunity to create a relationship with the pupil. The latter was reported to affect how they prioritised their work.

Relationships

This theme showed that teachers saw different relational values in the work around the assessment. When planning the work, they must decide which value should be prioritised. Prioritising was primarily described as three aspects that teachers dealt with in different ways: (1) impact on their ability to create relationships with individuals, (2) impact on their own attendance in their class and establishing relationships, or (3) giving the SEN teacher the opportunity to build a relationship with the pupils they will possibly teach later. For Åsa, the initial focus was on establishing a good relationship with the pupils when they started first grade:

For me, what is really important is to build a good rapport with the pupils and establish routines in the classroom. There are many occasions when someone else looks after the class and I feel that that ruin things a bit for me.

This was something that Gustav agreed with, so he and the SEN teacher he worked with tried to find a middle ground:

Sometimes I have, in my capacity as class teacher, been sitting with the pupils, but sometimes we do have help from the special needs teachers. This year, we started early in the term because we wanted to get to it early, and we felt we didn’t want to leave the class but let the special needs teachers perform the assessment for the early learning support guarantee, for the most part.

The teachers problematised different aspects about creating opportunities for relationships. The division of the task also needed to be done with respect to the pupil’s perspective, as well as a desire to establish a relationship. Johan said that the result of the assessment could be affected by the fact that the pupils were still quite young and not as familiar with the school’s staff:

Something to bear in mind, especially with first grade pupils, is that if they are a bit shy, for example, that needs to be taken into consideration. I feel that there are several new pupils that experience this, especially this early on in the first grade, so to send them to someone they don’t really—they hardly know me either—but they know the special needs teacher even less. So, from that perspective, it would be better if I conducted the support assessment.

Here, a central dilemma is illustrated. If the teacher carries out the assessment, they get a direct perception of the pupil’s knowledge, and an opportunity for social contact. On the other hand, the teacher needs to leave the relationship building in the classroom. If instead the SEN teacher carries out the assessment, the teacher risks not fully knowing the level of support a pupil will need to develop their number sense. The pupil may be negatively impacted by being judged by a person not known to them. On the other hand, the SEN teacher creates a social contact with the pupil. These were aspects of the dilemmatic space that teachers described having to navigate. They made choices to find possible ways forward – and achieved agency.

Temporal validity of the assessment result

If the SEN teacher carried out the assessment with all pupils, it meant they would not work with other forms of special education for most of the semester. Thereby, further dilemmas emerge regarding the follow-up work, this is exemplified by Richard:

During the assessment period, all our special needs education is put on hold while the special needs teacher conducts the support assessment.

Teachers in the study described it as a limitation that the SEN teacher’s time was completely used for the assessment. Because of the extensive work with the ELSG, there might be delays in delivering the support: ‘It takes a really long time before they get the support they actually need. To put it bluntly, one understands that the assessment needs to be done early, but the scheduling doesn’t work’ (Gustav). Richard, on the other hand, described advantages with the SEN teacher being the one who conducted the interview:

When we then sit down to plan for our pupils and to do the mapping of teaching needs and discuss various needs and how to divide our resources, we have someone who knows what the situation is like in the various classes. Otherwise, you’re very limited, ‘this is what it looks like in my class’, and so on. It’s good to have someone with that overview. That’s our reasoning anyway./ … /I have also noticed that it can turn into a bit of a competition to see who has the weakest pupils and who needs more help or support.

Richard talked about how implementation played a role in a pupil’s chances of receiving support. If the SEN teacher conducted the analysis of the result, then the support could be distributed more fairly. The division of labour was an active negotiating point in the teacher group regarding how resources were distributed. Maria reported that the fact that the SEN teacher conducted the interviews with her pupils led to dilemmatic aspects for her, as she had difficulty accessing the results, which in turn affected her follow-up work. She needed to actively seek out the SEN teacher to get the results:

We had our parent–teacher conferences during the same period as the assessment and we were very interested in finding out the results. We then sort of grabbed her to check ‘how did Pelle do’. But then we had to wait until she had created the report and reached a conclusion. And then we’d sit down together, and she would explain what we were looking at.

Maria chose to stay in her class but then had to deal with dilemmas that the SEN teacher had the information of the pupils’ results. When the SEN teacher conducted interviews to assess the pupils’ number sense, it resulted in consequences for pupils’ support. The consequences depended on the fact that the teacher did not have access to the pupils’ results or the fact that the results sometimes were no longer valid when the teacher received them. Gustav described how the pupils’ learning developed so quickly during the time the testing was going on that the first pupils’ results were no longer relevant when the last pupils were tested:

If you do the assessment with the first pupil, maybe they don’t have knowledge, but the last one does, as we have then had the opportunity to go through the material.

According to the teachers, the decision concerning who will carry out the assessment was vital. In the interviews it was consistently reported how identifying pupils need of support, paradoxically could become an obstacle for a pupil’s ability to actually receive that support. Some teachers achieved agency by taking over the assessment process themselves, to get the information needed to make an appropriate teaching plan, without involving the SEN teacher.

Discussion

There are previous studies that show that changing conditions for teachers’ work through reforms leads to limitations on teachers’ professional work (Avis 2005). In addition, when these new assignments are to be integrated into teachers’ ongoing work, the reform work creates new aspects in teachers’ ‘dilemmatic space’ (cf. Fransson and Grannäs 2013). In this study, we have highlighted professional dilemmas teachers in Swedish schools described that they encounter when implementing and fulfilling the work under the ELSG, and in what ways they achieve agency in this work. Reform work in education assumes that teachers have both the will and the capacity to implement the reform in line with its purpose (Biesta, Priestley, and Robinson 2015; Priestley et al. 2015). The purpose of the ELSG is situated in special education, as it should lead to early identification of pupils at risk of future mathematics difficulties. We believe it is possible to describe it as teachers achieving special educational agency to deal with these priorities.

The analysis shows that through the prioritisation in the decision of who will do the assessment interview with each pupil, the teachers achieve agency. This is evident in the overarching theme the right profession doing the right thing, where teachers expressed an ambition of taking responsibility for the reform. According to the ecological model of agency, three dimensions interact (the past, the present and the future). Achieving agency is described as the result of reactivated experiences and goals, together with cultural, structural, and material conditions at present (Biesta and Tedder 2007; Emirbayer and Mische 1998). When teachers described how work is organised and how they would like it to be organised, they related both to the context in which they worked, and to their future goals, consistent with the goals of the reform. They wanted to identify pupils who show a weak number sense early on and provide them with support. Within two dimensions of teachers’ agency, the present and the future (cf. Biesta and Tedder 2007), the teachers in this study described a ‘dilemmatic space’ (cf. Fransson and Grannäs 2013). This phenomenon affects how they achieve agency for developing pupils’ number sense. Furthermore, the dilemmatic space is framed by necessary priorities regarding cultural, structural, and material conditions in the current situation. The sub-theme control relates to how cultural conditions affect teachers’ opportunities to achieve agency. They described a desire to have both control over the individual pupil’s development of number sense and the regular work in the classroom with the whole class. Consequently, this endeavour led to teachers needing to prioritise their work in ELSG as teachers in grade one. The prioritisation is instigated by the reform. The intention to be the one in control is a prerequisite for agency and, thereby, an inherent prerequisite for teachers’ work. Also, having control over pupils’ early knowledge and arranging teaching in a way that relates to pupils’ prior knowledge is a cultural expectation teachers must live up to. This is especially evident in the ELSG.

Relationships and temporal validity of the assessment results are sub-themes that highlight structural conditions (cf. Biesta and Tedder 2007; Toom, Pyhältö, and Rust 2015) that appear in teachers’ descriptions as a prioritisation dilemma. In the assessment situation, the professional who carries out the assessment could create a personal relationship with the pupil. Establishing relationships was as valuable and relevant as finding out the pupil’s level of number sense. In addition, pupils’ opportunities for support were dependent on who conducted the assessment interview. If the SEN teacher conducted the interview, the SEN teacher could not simultaneously teach the pupils in need of support. Teachers described how a protracted implementation of the assessment interview can lead to the result no longer being a material resource. The assessment might not describe the pupils´ current level of number sense. Therefore, the purpose of the reform risks being compromised, since the assessment will form a basis for the future teaching the pupil will receive. Consequently, the long-term goal of creating better conditions for future mathematics development might be affected. This could be described as negative consequences assuming mathematics skills are seen as a prerequisite for success in school (Kilpatrick, Swafford, and Findell 2001).

Practical implications

The contextual conditions place great demands on a teacher’s ability to make decisions in order to find ways forward (i.e. achieving agency). This is performed by the teachers in this study primarily by making trade-offs around questions about where they think they do the most good. They also achieve agency in relation to a special education organisation that assumes that SEN teachers or teachers collaborate on teaching, either inside or outside the classroom. Cultural, structural, and material conditions affect how the teacher achieves agency to achieve their short- and long-term goals (Priestley et al. 2015). In conclusion, it is relevant to discuss whether teachers should conduct the assessment. One way of doing this might be to establish co-teaching between special needs teachers and classroom teachers (cf. Sundqvist, Björk-Åman, and Ström 2021). Being able to achieve agency is important if teachers are to make informed decisions and believe they can influence their work situation. Consequently, this is an example of work priorities creating a special education dilemma in teaching. In addition, there is a need to evaluate whether this is a matter of a failed policy or failure to implement the policy properly. Our study shown how teachers achieve agency when facing the direct consequences of implementing the policy. This means that further studies are needed to evaluate the reform, and its consequences.

Limitations

Some limitations of the study should be acknowledged. The analysis is based on interview data, and no performative data or observations have been collected. Therefore, we rely on the accounts of the participants, and as such, participants might describe a scenario as being more favourable to them. In addition, the study and the present findings are based on a small sample of Swedish teachers that may or may not be similar to other teacher groups. When valuing the generalisability of our findings, pattern recognition and context similarity (Larsson 2009) could prove useful.

Disclosure statement

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