The ambivalence of assessment — Language assessment of minority-language children in early childhood education and care

ABSTRACT

Language development in early childhood education and care (ECEC) has received increased policy attention in the past 20 years. Yet, few empirical studies have explored language assessment from the standpoint of ECEC teachers. Transnational organizations, such as the Organization for Economic Co-operation and Development (OECD), have increased their influence over national ECEC curriculums, stressing early intervention/readiness-for-school approaches to assist vulnerable groups in society. In contrast, the Nordic social pedagogy field perceives early intervention policies a threat to child-centered, playful approaches to learning. Based on interviews with 11 Danish and 11 Norwegian ECEC teachers in Copenhagen and Oslo municipalities, three main forms of ambivalence are identified: (1) ambivalence toward the ‘ready-for-school’ discourse, (2) ambivalence toward professional autonomy and discretion, and (3) ambivalence toward integration policy and the ideological code of ‘the standard child.’ The study recommends a more inclusive understanding of the implications of ‘adequate language proficiency’ and ‘school readiness.’

KEYWORDS:

• Early childhood education and care
• ECEC
• kindergarten
• language assessment
• minority-language children
• school readiness
• social pedagogy

Introduction

This article addresses how Early Childhood Education and Care (ECEC) teachers¹ approach the language assessment of minority-language children and how their everyday experiences are part of institutional relations and processes that transcend their local contexts. Immigrant and refugee families are increasing internationally, leading to a growing number of bilingual and/or minority-language children attending kindergartens (Castro and Prishker 2018). Concurrently, as Scandinavian populations have become more diverse societies, the kindergarten² has become a highly important integration arena for the youngest minority children (Tobin 2020; Bove and Sharmahd 2020). However, the recent Program for International Student Assessment (PISA) results raise concern, as Norway and Denmark (and the other Nordic countries) are among the Organization for Economic Co-operation and Development (OECD) countries with ‘the largest differences in favor of native-born students, after accounting for students and schools’ socio-economic profile’ (OECD 2019, 185). Early intervention and accountability have thus become central concepts in the policy field of ECEC. Although numerous studies have found that participation in ECEC is especially enriching for children from disadvantaged and/or minority families (Bakken, Brown, and Downing 2017; Vandenbroeck and Lazzari 2013), few have examined the integration of children from migrant backgrounds into ECEC (Bove and Sharmahd 2020) or ECEC teachers’ approaches to the language assessment of such children in practice.

During the past two decades, political interest in the transition from ECEC to primary school education has increased, and more responsibility for school preparation has been delegated to ECEC institutions, such as kindergartens (Blossing, Imsen, and Moos 2014; Imsen, Blossing, and Moos 2017; Christensen 2019). Thus, ECEC teachers have been given increased responsibility to detect and initiate special support for children with ‘inadequate’ development in the majority language.³ However, paradoxically, ECEC teachers play a minor role in the construction of professional standards (Havnes 2018). It is important to study these changes empirically, as the debates surrounding integration and standardized assessment policies are often politicized (Mausethagen, Prøitz, and Skedsmo 2018, 12).

The Norwegian and Danish ECEC sectors provide an interesting empirical comparison, as they have similar governance structures, as well as common pedagogical and historical origins in the ‘Nordic Model’ (Hännikäinen 2016, 1001; Mahon 2010), often contrasted with the more school-oriented Anglo-American preschool tradition (Hännikäinen 2016; Einarsdottir et al. 2015). The two countries resisted national standardized testing in education until the new millennium, disrupted by the ‘PISA shock’⁴ of 2001 (Tveit 2014, 2018). Therefore, language assessment in ECEC, primary, and secondary education is a relatively new phenomenon in Norway and Denmark (in contrast to countries such as Sweden).

However, in recent years, the two countries have taken different approaches to immigration and integration. While Denmark has Scandinavia’s most restrictive immigration policies, Norway has adopted an arguably more moderate approach (Hagelund 2020). This has strengthened the focus on standardized assessments of children’s language proficiency in Danish ECEC and on stricter language admission requirements for school entry in areas with high proportions of people from immigrant backgrounds (Danish Government 2018, 26). There are interesting differences between the two rather similar countries in terms of the standardized assessment in ECEC and the intersection between policies on integration and education. For instance, in 2018, Denmark implemented a high-stakes language screening test in immigrant dense and/or low socioeconomic neighborhoods. Children living in what the government deems ‘ghetto areas’ must pass a Danish language screening test to enter the first grade, or they are retained for a maximum of one year in a ‘kindergarten class’ on the school’s premises (Danish Government 2018, 26). Although Norway has no equivalent policy, both countries promote the use of standardized language assessment materials (TRAS (2011) and ‘Language assessment 3-6’ (2017)), which are designed primarily for monolingual Danish and Norwegian children, without considering appropriate measures for children with bilingual or multilingual backgrounds.⁵ This approach is contrary to government reports (Ministry of Education 2011) and international research on bilingualism and multilingualism, which are critical toward assessments that compare bilingual speakers against monolingual ideals (referred to as ‘a monolingual view of bilingualism’), as bilingual children’s language acquisition tends to develop differently from monolinguals (Drury 2013; Grosjean 1992; Henry and Thorsen 2018).

In this article, I ask: How do Norwegian and Danish ECEC teachers approach language assessment policies in practice? To answer this question, I present findings from interviews with 11 Danish and 11 Norwegian ECEC teachers concerning language testing policies in their everyday work with minority-language children. I find that ECEC teachers are ambivalent toward standardized testing in ECEC and argue that this relates not only to their own discretion in this area but also to an antagonistic discourse on the social mandate of kindergartens under changing political circumstances.

Literature review

Language assessment in ECEC is situated amid a scientific controversy between the supporters and opponents of standardized assessment practices (Vik 2017; Klem and Hagtvet 2018). The OECD has named these conflicting positions the ‘ready-for-school approach’ and the ‘social pedagogy approach’ (OECD 2006, 13;125). Variations of these terms have widely been taken up in research to describe the divide between the two approaches to curriculum development (Mahon 2010, 59; Einarsdottir et al. 2015; Wagner and Einarsdottir 2008). The ready-for-school approach has its roots in France and in English-speaking countries and stresses the importance of early intervention, academic skills, and measurable outputs, while the social pedagogy [Sozialpädagogik] tradition originated in the German Bildung pedagogy and is often associated with the Nordic region (Einarsdottir et al. 2015; Blossing, Imsen, and Moos 2014; Imsen, Blossing, and Moos 2017).

Research derived from the ready-for-school approach tends to stress early intervention to reduce socioeconomic disparities in educational outcomes and is dominated by contributions from economics and developmental psychology (Rege et al. 2018; Havnes and Mogstad 2011; Duncan et al. 2007). This tradition has a strong research focus on identifying which skills predict academic achievement and designing programs to enhance children’s development in these specific skill sets (Rege et al. 2018, 230; Havnes and Mogstad 2011; Mistry et al. 2010; Magnuson, Lahaie, and Waldfogel 2006; Rege et al. 2018, 230). Standardized assessment and documentation practices are valued as solutions to social inequality by ensuring equal quality across ECEC institutions and preventing local variations (Rege et al. 2018). The ready-for-school focus on children in ECEC is often positioned as conflicting with the Nordic ideals of holistic learning, child-centered perspectives, and child-initiated play (Børhaug et al. 2018; Hennum, Pettersvold, and Østrem 2015; Pettersvold and Østrem 2012; Blossing, Imsen, and Moos 2014; Imsen, Blossing, and Moos 2017).

In contrast, the social pedagogy research community has mainly focused on the harms of categorization (Pettersvold and Østrem 2012; Holm 2017; Houmøller 2018; Nilsen 2017a; Sjöberg 2014; Ehn and Petersen 2006; Klitmøller and Sommer 2015). A common criticism of standardized language assessment tools is that they generally reflect white middle-class values, which are presented as neutral norms of reference and place minority children at an unfair disadvantage (Nilsen 2017a, 60; Ehn and Petersen 2006; Palludan 2005; Houmøller 2018; Gulløv 2009; Slingerland 2017). Standardized assessment policies are also criticized for having an instrumental attention to children and childhood, implicitly focusing on children’s potential as human becomings rather than as human beings (Pettersvold and Østrem 2012; Qvortrup 1994).

Previous empirical research shows that political decisions at the transnational and national levels are increasingly intervening in the everyday work of ECEC teachers (Moss 2016; Nygård 2017; Christensen 2019; Kim 2018; Nilsen 2017b; Einarsdottir et al. 2015; Wagner and Einarsdottir 2008). Standardized materials, forms, and practices affect not only how teachers organize their everyday work with children but also how they view and relate to them in practice through the categorization in the materials (Nilsen 2017b; Houmøller 2018; Schmidt 2014; Vik 2018). However, few studies have empirically explored the standpoints of ECEC teachers, their approaches to language assessment, or how they navigate between the polarized ready-for-school and social pedagogy discourses.

Analytical perspectives

The analytical approach in this article is inspired by institutional ethnography (IE), a method of inquiry associated with the Canadian sociologist Dorothy Smith (2005). In IE, the researcher is especially interested in how people’s everyday doings are textually coordinated by others and how texts act as mediators of ruling. Hence, the ‘ruling relations’ between the actors in an institution are of primary interest. Through unpacking people’s interactions with texts in ‘text-reader conversations,’ the researcher can explore how ruling relations coordinate the social relations and activities in peoples’ everyday lives (Smith 2001). Large organizations are completely dependent on the infinite capacity of texts to copy and distribute the same message in local settings (Smith 2001, 165). Moreover, texts have the power to coordinate people’s activities by rewriting the local particularities of people into standardized, generalized representations of their work (Campbell 2006, 94). IE urges us to ‘set our predefined theories, concepts and understandings aside to avoid reproducing what we already know’ (Nilsen 2017b, 922). The actors interfering in people’s everyday activities are often concealed within texts, ideological codes, and institutional discourses, thus rendering the structures ‘faceless.’ An ‘ideological code’ can be understood as a schematic and replicable understanding that is incorporated into texts and discourses. An example is Smith and Griffith’s study of how the American education system was built on the ideological code of ‘the standard North American family’ or ‘the nuclear family’ and how this negatively affected the relationship between schools and parents in family constellations that did not fit the standard (Smith 1999, 159; Griffith and Smith 2005). The ideological codes in texts can thus reinforce narrow understandings of normalcy and deviance.

Materials and methods

This article draws on interview data from a study conducted in Copenhagen and Oslo from May 2019 until January 2020 and is part of an institutional ethnography of language assessment policies and practices in Norwegian and Danish ECEC. The teachers participating in the interviews were recruited from independent (selvejede) and municipal kindergartens in Copenhagen and Oslo. I interviewed 11 ECEC teachers in Copenhagen and 11 in Oslo from 14 different kindergartens: seven in Copenhagen and seven in Oslo. (For more information, please see Table 1. Characteristics of Participants) The informants work in kindergartens located in predominantly immigrant neighborhoods in both the inner city and suburban areas in Oslo and Copenhagen. I conducted all the interviews at the ECEC teachers’ workplaces; they all gave me a tour of the premises and let me introduce myself to the children and staff. Four out of the 18 interviews were group interviews (two with two informants and two with three). Both types of interviews contribute to the empirical material in different ways. During the one-on-one interviews, it was easier to lead the conversations and explore the topics I wanted to touch on, while the conversations more often drifted thematically when there were two or three participants. On the other hand, this was what made the group interviews so useful, as the ECEC teachers often felt at ease discussing topics with their colleagues and sometimes disclosed more new information and ambivalent feelings than in the sometimes more ‘to the point’ one-on-one interviews.

Table 1. Characteristics of participants.

Informants		Kindergarten	Type of interview	Kindergarten ownership	Kindergarten size by enrolled children	Years of experience	Country
1	Mona	Sun	Group	Municipal	50–99	25+	Denmark
2	Edith	Sun	Group	Municipal	50–99	16–20	Denmark
3	Jakob	Saturn	Individual	Independent	50–99	6–10	Denmark
4	Aisha	Jupiter	Individual	Municipal	19–50	11–15	Denmark
5	Merete	Mars	Individual	Independent	50–99	6–10	Denmark
6	Casper	Venus	Group	Independent	50–99	25+	Denmark
7	Camilla	Venus	Group	Independent	50–99	-	Denmark
8	Elisabeth	Earth	Individual	Municipal	19–50	0–5	Denmark
9	Patrick	Europa	Group	Municipal	50–99	6–10	Denmark
10	Karen	Europa	Group	Municipal	50–99	21–25	Denmark
11	Anne	Europa	Group	Municipal	50–99	0–5	Denmark
12	Caroline	Moon	Individual	Municipal	50–99	16–20	Norway
13	Roger	Neptune	Group	Municipal	50–99	-	Norway
14	Ruth	Neptune	Group	Municipal	50–99	21–25	Norway
15	Christina	Neptune	Group	Municipal	50–99	11–15	Norway
16	Tone	Uranus	Individual	Municipal	50–99	16–20	Norway
17	Michael	Uranus	Individual	Municipal	50–99	25+	Norway
18	Eric	Uranus	Individual	Municipal	50–99	0–5	Norway
19	Harald	Pluto	Individual	Municipal	50–99	25+	Norway
20	Marie	Mercury	Individual	Municipal	50–99	16–20	Norway
21	Peter	Ceres	Individual	Municipal	50–99	6–10	Norway
22	Turid	Eris	Individual	Municipal	50–99	25+	Norway

During the interviews, the ECEC teachers were asked about their approach to language work and assessment in their everyday interactions, especially in relation to minority-language children (their ‘work knowledge’). They were encouraged to elaborate on their practices and experiences and to provide detailed explanations. I only recruited ECEC teachers from independent (selvejede) kindergartens in Copenhagen as these are subject to the same language assessment regulations as the municipal ones. The Norwegian counterpart of independent kindergartens – called ‘ideal kindergartens’ – are not obliged to follow Oslo’s language assessment policies and were therefore not included in the study.

The study has been approved by the Norwegian Center for Research Data (2020) and complies with the Norwegian National Guidelines for Research Ethics in the Social Sciences, Humanities, Law and Theology (2016). The names of the people and places have been anonymized. I transcribed the sound recordings from the interviews verbatim and coded the material sentence by sentence using Nvivo. I first coded the material inductively by densifying the meaning construction within each paragraph, followed by a deductive phase in which I focused on the new codes that emerged from my analysis of the codes from the first phase, such as ‘ambivalence’ and the ECEC teachers’ pragmatic approach to language assessment (Tjora 2012). Furthermore, I coded the material by formal and informal stopping points in an assessment process, from a child entering kindergarten until leaving for school. Moreover, I coded the institutions, texts, and actors involved at each stopping point. A stopping point could be a meeting with parents or completing an application for extra resources for a child.

The ECEC teachers’ work knowledge directed me to the institutions, discourses, and values shaping their everyday interactions with minority-language children and their families. My analysis was inspired by DeVault and McCoy’s three-stage process for investigating ruling relations. First, I identified an experience, such as a majority of ECEC teachers experiencing ambivalence toward performing language assessments because several minority-language children in their kindergarten had failed test questions referring to cultural expressions they did not recognize (contrary to children with monolingual Danish or Norwegian backgrounds). Second, I identified the institutional processes that shaped that experience, such as assessment policies, discourses, and ideological codes. Third, I investigated those processes more closely to analyze how they influenced the experience (DeVault and McCoy 2006, 20).

Copenhagen and Oslo were chosen as the starting points for this inquiry, as policy implementation in these capital cities is highly influential in the respective national contexts.⁶ The Copenhagen municipality does not mandate that all children be screened for language proficiency, but the Day Care Act states that all three-year-olds should receive a language assessment ‘in cases where there are linguistic, behavioral or other circumstances that require the child to receive language stimulation’ [my translation] (The Day-Care Act 2018, §11). The testing of all children in an area is decided at a higher administrative level of the city district to which the kindergarten belongs. In practice, all children enrolled in the Danish kindergartens in my study were to be screened with the ministry’s own ‘Language assessment 3-6’ [Sprogvurdering 3-6] (The Ministry of Children and Education 2017). In Oslo, ‘the Oslo standard’ requires all children to be ‘systematically observed,’ but does not specify how (Oslo City Council Administration 2019). No Oslo kindergarten in my study screened all children, but the Resource Centers in their local city districts require them to use ‘TRAS’⁷ (Espenakk et al. 2011) to assess children’s language development for specialist referrals or applications for extra resources.

Results

The majority of the ECEC teachers participating in this study were mostly practically oriented, and the most common objections to assessment policies concerned the lack of time and resources. The ECEC teachers espoused a pragmatic approach and an ambivalent attitude toward assessment in their comments on everyday practices and adjustments to their work to meet managerial demands. They all strongly valued ‘early intervention’ and spoke of the importance of starting ECEC at an early age at the same time as advocating for the social pedagogic approach of child-centered activities and play-based learning. Both Danish and Norwegian ECEC teachers continually switched between both the social pedagogy and ready-for-school discourses in their approach to language work, even though these are portrayed as opposing traditions in the research literature. In both countries, the ECEC teachers emphasized that how the assessments were conducted was more important than whether all children’s language was assessed or not.

The Danish and Norwegian ECEC teachers had quite similar experiences of working with minority-language children and their families. The largest difference was the status of the assessment tools in the two countries. In Norway, the freedom of method was interpreted in practice to mean that only children with suspected learning difficulties or other diagnoses should be assessed. In Denmark, the assessment materials formed a screening test integrated into the municipality’s apps – Copenhagen Children and Copenhagen Parents – and hence were interpreted in practice as meant for all children.

I apply the term ambivalence to describe the ECEC teachers’ experiences, as they continuously navigate uncertainty caused by conflicting values and messages from various actors and discourses as well as uncertainty regarding their own professional discretion. In the following sections, I present the three most prominent types of ambivalence amongst the ECEC teachers: ambivalence toward the work of assuring school readiness, ambivalence toward professional autonomy and using discretion, and ambivalence toward integration policy and the ideological code of ‘the standard child.’

Some ECEC teachers were more ambivalent than others concerning all three aspects, but they usually had the main tendencies in common. The forthcoming sections focus on the similarities between the Danish and Norwegian ECEC teachers’ approach to assessment, as these similarities were striking despite different political and national contexts.

Ambivalence toward the work of assuring school readiness

This type of ambivalence is associated with research from the ready-for-school tradition and the influence of international organizations, such as the OECD, on the ECEC curriculum. The governance of the accountability system of ECEC allows schools to determine what is ‘adequate,’ and kindergartens are held accountable for the school’s assessment of children’s ‘school readiness.’ The sanctions for the lack of school readiness vary between countries. In Norway, it is vastly uncommon to hold children back in kindergarten or have them repeat grades. In contrast, Danish children deemed ‘unready for school’ are regularly retained in kindergarten or must repeat preschool. All the Danish ECEC teachers reported schools requesting that they ‘take a child back’; however, this is not always possible, as the child’s previous place in the kindergarten may already be filled. This left the Danish ECEC teachers especially worried that their children’s language development would be ruled ‘inadequate.’ At the same time, both Norwegian and Danish ECEC teachers reported that school and kindergarten institutions had conflicting expectations concerning school readiness and adequate language proficiency.

Despite this, the teachers pragmatically accepted the schools’ academic standards for school readiness, but at a professional level, they disagreed to some extent about what school readiness should entail, as the ECEC teachers were more focused on practical independence skills than the schoolteachers were. The kindergartens in both the Oslo and Copenhagen municipalities are obliged to submit a standardized transition form prior to a child starting school, in which they have to account for the child’s language proficiency in Norwegian or Danish, respectively (Child and Youth Department Copenhagen Municpality 2020; Oslo City Council Section for Childhood and Education 2020). The Oslo-based ECEC teachers (Terum and Molander 2008) reported that schools hold the kindergartens responsible for reporting language proficiency ‘accurately’ (i.e. according to the school definition) when children transition to school. Below, ‘Emilie’ recounts an experience of communication with the local school in relation to school transition:

We received an anonymized list (from the school) of children arriving from this kindergarten that ended up needing ‘specific language support’ [in the first grade]. [The list identified] whether we had marked them as needing special language support or not [in the transition forms], and whether they had passed the NISK test (the school’s language proficiency test). We were additionally asked to provide documentation on why we had not reported on all the children [who later failed the school’s test] and later had been granted special education. Emilie, 27, Pedagogic Leader, Oslo.

This was not a unique incident. Other ECEC teachers that I interviewed in Oslo reported similar experiences of schools contacting them to complain about kindergarten documentation and ‘underreporting’ of children with ‘inadequate language development.’ The ECEC teachers often commented that they did not see the children’s language development as inadequate in the sense that the child needed special pedagogic help; rather, they often thought that the child needed more language stimulation in Norwegian. However, the ECEC teachers in both countries also agreed that the standardized assessments made it easier to communicate concerns regarding children’s language development to schools, other institutions, and parents. Standardized assessment and documentation could thus relieve the ECEC teachers in the sense that everything was documented in a way that demanded attention from other actors.

(…) sometimes, if we don’t have it on paper … for example, if I just go and tell mom or dad that ‘your child lacks some words and we should do this or that … ’ [She gesticulates that the parents just wave her away and laugh at her] [The parents answer:] ‘No, he will learn in due time. It will resolve itself with time.’ But when I have it on paper and they can see the red color, then it gets serious; then it is important. And it is actually good to have it, so they can see … And it is the same with the municipality. Because when you send it, it’s official. Aisha, 47. Pedagogue, Copenhagen.

The ECEC teachers in both countries said that the results from the standardized assessments were especially useful when they had difficulty convincing parents to take their concerns seriously. Hence, the texts can be utilized to ensure that a child receives the help the ECEC teachers believe is needed.

Ambivalence toward professional autonomy and using discretion

Ambivalence toward professional autonomy is related to the ECEC teachers’ experiences of their discretion to decide whether a child has adequate language development. The struggle for professional autonomy concerns politics, as autonomy is important for protecting professional status, but also to ensure pedagogical quality by providing teachers the agency to adjust their practice to meet the needs of individual children and contexts (Pettersvold and Østrem 2018b; Terum and Molander 2008; Mausethagen and Mølstad 2015). This ambivalence was most prominent among the Norwegian ECEC teachers, as their Danish counterparts relied heavily on screening devices, somewhat relieving their uncertainty over determining children’s language proficiency. However, the assessments could not relieve the ECEC teachers’ concern over whether inadequate language proficiency was caused by a lack of stimulation in the majority language or by a learning difficulty.

When I asked how the ECEC teachers determined adequate language proficiency for standardized forms for transition from kindergarten to school,⁸ many Norwegian teachers claimed that ‘it is just something you know.’ Most Norwegian teacher-based concerns over a child’s language development stem from factors other than assessment results, indicating that the teachers trust their professional discretion without assistive devices. They only assessed children when they were already concerned. At the same time, many Oslo-based ECEC teachers reported frustrating experiences of determining adequacy without assessment tools. Turid was worried about the consequences of heightened segregation in the Oslo municipality and the related effects on her and her colleagues’ discretionary abilities:

I have said it the entire time to the managers here: I think many of the children we send to the schools have too poor Norwegian skills. (…) I wish that I had more Norwegian children here, so you could compare more, because we have become blind here (in this kindergarten) since we have so few (majority language-speaking children) to compare with. Turid, 55. Pedagogic leader, Oslo.

The problem of determining ‘adequate language proficiency’ was also recognized by the ECEC teachers in the Neptune kindergarten. In the following excerpt, Ruth and Christina discuss their shock when visiting another kindergarten and realizing that they had completely different perceptions of what adequate language proficiency was, compared to the staff at a neighboring kindergarten. This became apparent when two children from different kindergartens were compared to one another during a visit:

Ruth:

You’re in this [your own kindergarten] all day. And it is pretty frightening, because you assume that the language development level [of the children in your own kindergarten] … when you hear them speak, you think ‘Oh, this is good,’ but it’s not.

Christina:

You kind of ‘hear yourself blind,’ you get so used to … We discussed this: ‘God, how actually proficient are the children we regard as [adequate] in our kindergarten … how well do they actually speak?’ […] It was quite an epiphany ‘How well does this child do … who we think is a good language speaker, how good is she actually compared to where she is supposed to be?’ And then you really see the discrepancy, two children the same age, one was actually older, [who was] considerably worse at language comprehension and oral presentation, when we actually [previously] thought the child had ‘adequate proficiency.’ Ruth, 55. Manager, and Christina, 36. Pedagogic leader, Oslo.

Thus, basing determinations of language development solely on discretion creates variations in the ways Oslo kindergartens determine adequate language proficiency and when to raise concern. Several teachers noticed that every kindergarten constructs a local definition of adequacy based on its own group of children. These findings suggest a tendency toward the child with the strongest language proficiency anchors the standard of adequacy for the remaining children (Kahneman 2011). Both the Danish and Norwegian ECEC teachers also reported that they were sometimes surprised by the results of language assessments. For example, an outspoken child might be mistaken as having a stronger language proficiency than was the case or an introverted child as having poorer language development.

Ambivalence toward integration policy and the ideological code of ‘the standard child’

This form of ambivalence relates to the political debate over immigration, integration, and the hierarchical relations between the majority and minority groups in society. Despite different national strategies on language assessment and immigration, both the Norwegian and Danish ECEC teachers experienced conflicting feelings regarding the role of kindergartens as integration arenas and which values were represented in the assessment/screening tools and materials. The teachers from both countries all underlined the importance of focusing on the positive aspects of multiculturalism⁹ in the daily life of the kindergarten. They sang popular children’s songs in the children’s home languages, decorated the kindergarten’s interior with words in the children’s home languages, and involved parents in finding music to play during their daily activities to foster a sense of belonging and pride amongst the children. The minority-language children attending ECEC and their families are often already in a vulnerable social position because of the intersectionality between immigrant background and low socioeconomic status (Thorsen 2021; Galloway et al. 2015; Ottesen et al. 2018). Moreover, the turbulent political climate over immigration and integration in the Nordic countries makes the topic of the integration and assessment of minority-language children rather sensitive.

Both the Norwegian and Danish ECEC teachers pointed out several examples of assessment materials based on the notion of the ‘standard child’ from the majority group, as monolingual, white, and middle class. This is an excerpt from an interview with the Danish pedagogue Mona, 45. I have named the predominantly immigrant area in Copenhagen ‘Broen’ and the predominantly white middle-class area ‘Borgen.’

Mona:

(Silently showing me a picture in the testing material, pretending playfully that I am the child being tested).

Interviewer:

It’s a tie, right?

Mona:

Yes, it’s a tie. There are no children here that (answer that question correctly). We are situated in Broen … I live in Borgen, where they know what it is. Because dad uses a tie and works in a bank, (the children) know well what it is, right. But here, no child has ever answered that one correctly. They don’t know what it is (…), so you lose some points here (on the tie question) if you live in Broen. But you don’t if you live in Borgen.

The second example comes from Norway. Tone was working with a tool called SPROFF, through which both the minority-language children’s language development is assessed and an intensive standardized course built around different topics is followed each week.

I remember we had the topic of camping, and there were tents, sleeping bags, a cabin and mountaintops, and none of these children had any relationship to any of it. Tone, 63. ECEC teacher, Oslo.

Both the question about the tie and the references to sleeping bags and cabins in the mountains reflect Nordic culture and/or middle-class culture, instances in which the children’s inability to answer possibly has just as much to do with their lack of knowledge of Nordic culture as their language development (Bourdieu 1984; Bourdieu and Passeron 1977; Nilsen 2008).

Despite the ECEC teachers’ frustration with the materials, their concerns for the minority children’s life chances exceeded those about the negative effects of standardized assessment. They were more concerned that the school’s heightened academic focus and Oslo and Copenhagen’s increasing segregation impaired the minority children’s chances of keeping up with their monolingual majority peers when starting school. The minority-language children often started kindergarten at a later age than the majority children, and the teachers underlined the importance of early intervention:

We have a girl in my division; she came right before she turned three years old. When she came, she was at the level of a one-and-a-half year old in motor skills, socially, and in every other aspect … and if she had attended nursery from when she was one, she would have come a bit further … at a developmental level, in both motor and linguistic skills. Merete, 40. Pedagogue, Copenhagen.

This was a common perception amongst both Norwegian and Danish ECEC teachers. Many teachers wanted kindergarten to be compulsory from the age of one year and for the municipalities to remove the free choice of kindergarten to prevent the growing segregation. Simultaneously, the ECEC teachers felt ambivalent toward these types of policies as they are usually associated with other more restrictive immigration and integration policies that they do not support.

Discussion and conclusion

In this article, I asked the following research question: How do Norwegian and Danish ECEC teachers approach language assessment policies in practice?

I found that ECEC teachers were mostly pragmatic in their approach to language assessment, but they were ambivalent in relation to language work and the assessment of minority-language children. These are important topics to discuss critically as they can renew and differentiate our understanding of ECEC teachers’ standpoints and approaches.

I identified three sources of ambivalence that are most prominent among the ECEC teachers: (1) ambivalence toward the ready-for-school discourse, (2) ambivalence toward professional autonomy and the use of discretion, and (3) ambivalence toward integration policy and the ideological code of ‘the standard child.’ Their ambivalence relates not only to their own discretion but also toward an antagonistic discourse on the social mandate of kindergartens under changing political circumstances. The types of ambivalence relate partly to different actors: researchers in the field, professionals, and politicians.

The findings in this article differ somewhat from those in the previous research, which has focused more on the tension between professional ethos and neoliberal discourse (Hennum, Pettersvold, and Østrem 2015; Pettersvold and Østrem 2018a, 2012) and less on ECEC teachers’ ambivalence toward the social pedagogic rejection of standardized assessment and documentation policies. As the previous research shows, the ECEC teachers from both countries underline that how and why standardized language assessments are performed are the most important factors (Klem and Hagtvet 2018; Vik 2018). Their practical and pragmatic orientation toward assessment is congruent with previous research on ECEC and secondary school teachers’ approaches to standardized assessment policies in practice (Nilsen 2017a; Mausethagen, Prøitz, and Skedsmo 2018, 2021).

Similar to the previous research findings, not only is language assessment being standardized but also other pedagogical everyday practices of ECEC staff (Christensen 2019; Houmøller 2018; Nilsen 2017a), such as teachers’ interactions with parents, their views on children’s language proficiency, and their efforts to achieve school readiness. Despite the Danish and Norwegian ECEC teachers’ rather similar approaches to assessment and everyday language work, the standardized assessments materials received a higher status in Copenhagen than in Oslo.

In Copenhagen, the government’s language assessment was designed as a screening tool, and since it was integrated into the municipality’s mandatory kindergarten app, it received a higher status in the Danish ECEC teachers’ everyday work. While in Oslo, the most common assessment material (TRAS) is constructed as a form that can be completed by ECEC teachers without the children’s presence (nevertheless, the ECEC teachers often test children face-to-face if they are unsure).

Furthermore, based on the findings concerning the ECEC teachers’ approach to language assessment, I discussed how transnational discourses and agencies are involved in their language assessments of minority-language children. The problem of determining school readiness and whether a child has adequate language development is linked to translocal discourses such as ready for school and early intervention, which are produced by international interest organizations, such as the OECD (2006, 2018), and dominant research on ECEC in economics and developmental psychology (Havnes and Mogstad 2011; Bettinger, Hægeland, and Rege 2014). The ECEC teachers’ experiences of ambivalence are connected to a larger discussion regarding the social mandate of kindergartens in changing the political circumstances of integration in the Nordic countries. However, there is also a struggle between ECEC and the teaching profession to determine responsibility and accountability for minority-language children’s language proficiency. The ECEC teachers’ concern about the minority-language children’s school readiness is related to the education system being based on the perception of the standard child as monolingual and belonging to the majority culture (Gulløv 2009; Sønsthagen 2020; Grosjean 1992). This perception can also be referred to as an ideological code (Smith 1999).

Although the findings are not generalizable in a quantitative sense, they have transferrable value as they touch upon topics that transcend national borders. They illustrate blind spots in the assumptions of integration and professional discretion on which assessment tools and policies are built. The simplified worldview presented by both the ready-for-school and social pedagogy discourses hides the complexity of the everyday activities and relations that structure ECEC teachers’ language work with minority-language children and their families. Without understanding this complexity, it is difficult for policy makers to provide integration and assessment policies that meet the needs of minority-language children.

Implications

The risk of assessment materials being based on the ideological code of a majority standard child not only weakens the reliability of the assessment tools themselves, but the neutralization of ‘the standard child’ as ‘the school-ready child’ can have more serious implications. When the testing materials are given the status of an objective measure, they can recreate and reinforce social inequality between minority and majority children in ECEC and society as such (Nilsen 2017b). However, standardized assessment and documentation practices can also help to prevent cognitive bias (see, for example, Kahneman (2011) for an introduction to the literature on heuristics and biases) and provide a more effective way of legitimizing and communicating concerns about a child between the ECEC institution and other actors. Therefore, the discussion regarding language assessment in ECEC should not center on ‘either/or’ but rather consider and reflect on the knowledge and values on which the materials are founded. It should move toward a more inclusive understanding of the implications of the terms ‘adequate language proficiency’ and ‘school readiness’ as well as the implementation of assessment and documentation in ECEC for both majority- and minority-language-speaking children.

Disclosure statement

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

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Notes

1 The term ‘ECEC teacher’ refers to professionals with either Danish pedagogical training or Norwegian kindergarten teacher training.

2 I apply the term ‘kindergarten’ to refer explicitly to the characteristics of the Nordic model of early childhood education and care, in which the Fröbel tradition is prominent. This is the terminology used by ECEC teachers themselves to describe their workplace (‘Børnehave’/‘Barnehage’ in Danish and Norwegian, respectively).

3 Research suggests that children from bilingual backgrounds tend to be overidentified with language development issues due to educators’ lack of appropriate developmental expectations (Bedore and Peña 2008)

4 The ‘PISA shock’ refers to reactions in the wake of the publication of the first test results from the Programme of International Student Assessment (PISA) in 2000. The ‘shock’ was that many national states were negatively surprised by their own population’s low test scores compared with other countries(Tveit 2014, 2018).

5 TRAS was not developed with bilingual children in mind, but it was later added a similar assessment scheme, in which the only difference is the removal of the age indications present in the coloring scheme for monolingual children(Espenakk et al. 2011).

6 Norway and Denmark have highly subsidized ECEC, and in the age group 3 to 5, 97.1% of Norwegian children and 88.6% of Danish children are enrolled in kindergarten (Drange and Telle 2018; Glavind and Pade 2018).

7 ‘TRAS’ stands for ‘Tidlig Registrering av Språkutvikling’ in Norwegian, meaning ‘Early Registration of Language Developement’ (Espenakk et al. 2011).

8 The yes/no question of whether ‘the child has adequate language proficiency’ has since been removed in the revised edition (May 2020) of the Oslo standard (Oslo City Council Section for Childhood and Education 2020).

9 Multiculturalism can very simply be understood as ‘a generic term for the ensemble of policies introduced with the combined goals of recognizing diversity, fostering integration and producing/maintaining equality’ (Taylor 2012, 415).