Decision-making factors for group organising in Swedish preschools

ABSTRACT

This study explores preschool teachers’ considerations when organising children into subgroups in Swedish preschools. Grounded on interactionist perspectives and Bronfenbrenner’s ecological systems theory, the study hypothesises that preschool teachers’ practice of organising children into subgroups is embedded in and shaped through dynamic interactions with all actors and levels of the entire preschool system. Based on this theoretical framework, a hypothetical model has been specified that brings together variables related to children’s characteristics, preschool teachers’ interpersonal relationships and working conditions as well as preschools’ organisational aspects. The sample consists of 698 preschool teachers from different preschools in 46 municipalities in Sweden who completed an online questionnaire. By applying Confirmatory Factor Analysis technique in evaluating the hypothesised model, the study’s objective was to identify factors that determine preschool teachers’ decisions when organising children into subgroups. The analysis identified a three-factor model where two residual but distinct factors, namely an interrelational and a structural factor, were nested under one general factor. The results suggest that organising children into subgroups in preschool is partly a collectively negotiated practice and partly closely connected to the complexities of preschool classrooms’ actual conditions.

KEYWORDS:

• Preschool teachers
• group organising
• subgroups
• ecological systems theory
• confirmatory factor analysis

Background

This study explores preschool teachers’ considerations of grouping practice in Swedish preschools. By applying confirmatory factor analytical (CFA) method, the purpose of this study is to identify factors that determine preschool teachers’ decisions when organising children into subgroups. The research literature on preschool classroom quality and outcomes underpin that high quality preschool provisions have a positive impact on children’s cognitive and social development (La Paro et al. 2009; Pianta et al. 2005; Sylva et al. 2010). Considering that children’s experiences in preschool classroom settings and activities have been associated with academic performance and behaviour several years later (Booren, Downer, and Vitiello 2012; Mashburn et al. 2008), classroom organisation is an important feature in preschool quality. Classroom organisation often refers to strategies that preschool teachers use intentionally to maximise opportunities for learning and well-being by establishing and maintaining order, effectively engaging children, knowingly choosing activities and productively making use of children’s time in preschool (Curby et al. 2009; Emmer and Stough 2010; Pianta et al. 2005). Providing small group activities and thereby organising the whole preschool group into subgroups is an important quality aspect often associated with preschool teachers’ professional competence to structure the preschool day in creating challenging environments for children’s well-being, learning and development (Sheridan, Williams, and Pramling Samuelsson 2014).

Despite the research demonstrating that children’s learning and development flourishes in a balance of both structured and unstructured preschool classroom settings (Siraj-Blatchford et al. 2008; Sylva, Ereky-Stevens, and Aricescu 2015), researchers have found that preschool teachers often view teacher-managed settings, such as working in subgroups, as appropriate for academic instruction (Cabell et al. 2013; Early et al. 2010; Leggett and Ford 2013). Other studies have also emphasised the importance of group arrangements made in preschool in supporting young children’s emotional interconnectedness (van Oers and Hännikäinen 2001), sense of community (Koivula and Hännikäinen 2017; Rogoff 2003) and friendship (Corsaro 2003).

To date, little attention has been given to different factors that may affect preschool teachers’ choices of grouping practices. This research gap, coupled with the increased emphasis on children’s academic learning and the importance of preschool teachers’ professional competence in teaching and supporting conditions for children’s learning and development, necessitates a closer investigation of this topic.

The current study endeavours to contribute to knowledge on preschool teachers’ grouping practice. This practice is, in the study, considered as preschool teachers’ organisational and pedagogical strategy to arrange children into subgroups throughout the day. In the Swedish preschool context, there is limited research on grouping practices in preschool and even less on factors that may influence preschool teachers when they make decisions on how to organise children into subgroups. In line with this, the research question is:

What are the latent factors that can be identified by preschool teachers’ expressed considerations when they organise children into subgroups?

Grounded on interactionist perspectives (Bergman, Magnusson, and El Khouri 2003; Bronfenbrenner 1979, 1986) and ecological systems theory (Bronfenbrenner 1979, 1986), the study hypothesises that preschool teachers’ strategies when organising children into groups are not only a feature of their pedagogical beliefs and intentions but also an attribute of the entire system imposed by the complexities of the classrooms’ actual conditions (Bronfenbrenner 1979, 1986; Urban and Dalli 2012). Based on this rationale and by applying CFA, this study uses a set of manifest indicators representing multiple dimensions for organising children into subgroups in preschool. These indicators are related to children’s characteristics, preschool teachers’ interpersonal relationships and working conditions as well as preschools’ organisational aspects.

The Swedish preschool context

In Sweden, preschool offer educational group activity for enrolled children between the ages one to five years. Preschool has been a part of the educational system since 1996, and the preschool curriculum came in force in 1998 (Swedish National Agency for Education 2010). Based on a holistic perspective of a child’s learning and development, the concept of educare is central to the Swedish preschool practice in the sense that play, learning and care are interconnected (Broström 2006; Johansson and Pramling Samuelsson 2001). In 2010, preschool became a distinct form of school regulated by the New Education Act (Ministry of Education and Research 2010). The preschool curriculum was revised in 2010, 2016 and 2018 with a strong emphasis on both content areas such as language, mathematics, science, and technology and documentation of preschools’ pedagogical work and didactic issues. In 2011, a new preschool teacher education programme with a clear didactic perspective was introduced. In the revised curriculum and the new education programme, preschool teachers’ teaching responsibilities have been strengthened. These changes in Swedish preschools highlight a shift in preschool pedagogy towards a more learning-oriented approach (Sheridan et al. 2011).

Besides these changes in the Swedish preschool pedagogical approach, the increasing proportion of children attending preschool from an early age has generated preschool teachers’ efforts to organise their daily work in preschool classrooms through different group arrangements (Sheridan, Williams, and Pramling Samuelsson 2014; Williams, Sheridan, and Pramling Samuelsson 2018). The child group in preschool is thus seen as a fundamental and active component in children’s possibilities to develop and learn from an individual and collaborative perspective (Swedish National Agency for Education 2016). Preschool teachers, therefore, need to have the professional knowledge and competence to organise content and group activities to maximise children’s opportunities to learn and develop (Sheridan et al. 2011). In sum, the increased demands for more knowledge-oriented preschools and the large number of children enrolled in preschool have increased expectations on preschool teachers’ professional competence to intentionally structure the preschool day into various groups in order to provide a rich learning environment and meet children’s needs.

Research on classroom organisation and grouping practices in preschools

The recent focus on young children’s academic learning has resulted in numerous studies in the Early Childhood Education and Care (ECEC) field using observational instruments to measure and evaluate the quality in preschool classroom settings. For example, the Classroom Assessment Scoring System (CLASS; Pianta, LaParo, and Hamre 2008), widely used in the ECEC research field in the United States and validated in Finnish preschools by Pakarinen et al. (2010), is a structured observation of preschool classroom practices designed to measure the quality of teacher–child interaction and classroom functioning across three domains: emotional support, classroom organisation and instructional support. A high level of quality in these domains has been found to contribute to children’s social and cognitive development (Pianta, LaParo, and Hamre 2008; Salminen et al. 2012). As demonstrated by the research associated with CLASS, well-organised classrooms and effective classroom management are associated with children’s motivation (Pakarinen et al. 2010) and greater engagement in learning activities (Rimm-Kaufman et al. 2009; Vitiello et al. 2012).

The Early Childhood Environment Rating Scale (ECERS; Harms, Clifford, and Cryer 1998) is another internationally well-documented evaluation scale of preschool classroom quality specifically designed to measure aspects of structural and process quality, such as the classroom’s physical environment, materials, routines, indoor and outdoor learning activities and programme structure (e.g. free play, group time). Several studies underpinned by ECERS have pointed out that in high-quality preschools, children’s development and learning were associated with preschool teachers’ awareness and professional competence to structure the whole day with curriculum activities, working in subgroups and providing teacher-planned and – initiated focused group work in balance with child-initiated play (Sheridan 2001; Sylva et al. 2010).

There is a growing body of research exploring time management in various settings, activities and teaching interactions with mixed results regarding the academic benefits of time children spent in various types of activities. One example is a study by Early et al. (2010), who used classroom observational data of 2061 children in 652 preschool programmes to explore the amount of time children spent in child-directed activities (free play), teacher-assigned activity settings (whole group, small group, individual) and meals/routines. While they found that a preschool day was approximately equally divided among these domains, the amount of time spent in whole group activities, typically teacher structured, was negatively correlated with classroom process quality. A profiling study by Fuligni et al. (2012) exploring how teachers structure daily routines in preschool settings identified a high free-choice pattern and a structured-balanced pattern. Being in a classroom with structured-balanced daily routines in which children spent equal proportions of time in both child-directed and teacher-directed activities was associated with greater opportunities for children to engage in academic activities and scaffolding teacher–child interactions.

The organisation of time in preschool, the structure of professional practices and daily life for both children and staff have also been the focus of several studies in the Nordic countries. Kjørholt and Seland (2012), for example, criticised the organisation of preschools in Norway that utilise extended working teams of preschool teachers and a large number of children in open spaces. The authors claimed that this alternative form created by a shift from traditionally organised small and fixed groups to larger and multifaceted groups of children, has a significant effect on children’s well-being and learning, exposing them to extended social relationships in an unconstructive way. It also has a negative impact on preschool staff’s working conditions, requiring increased effectiveness and professional competence to organise everyday practice and fulfil curriculum aims. Similarly, Alvestad et al. (2014) studied the experiences of preschool staff working with toddlers in large groups and with various group organisational models in Iceland, Norway and Sweden. Based on focus interviews, the study found that working with small groups of children is an effective organisational strategy associated with better atmosphere in groups, which increased children’s participation and concentration in learning activities and provided more opportunities for interaction between teachers and children.

In Sweden, studies from a recent research project with a mixed method design investigating the impact of group size on children’s affordances in preschool (Williams, Sheridan, and Pramling Samuelsson 2016, 2018), found that dividing the children into smaller groups during different parts of the preschool day was the preferred practice to make the entire group more manageable for teachers (Pramling Samuelsson et al. 2016; Sheridan, Williams, and Pramling Samuelsson 2014). While these studies found that the organisation of children into small groups takes place during a limited part of the preschool day, the preschool teachers in these studies expressed that the organisation of children into groups was a functional way to conduct their work in line with curriculum intentions. Based on interviews of 24 preschool teachers, Sheridan, Williams, and Pramling Samuelsson (2014) explored Swedish preschool teachers’ perspectives on organisational conditions for children’s learning in relation to group size. They found that a range of interconnected factors, including children’s personal characteristics (e.g. age, gender and special needs), structural and environmental aspects of preschool (e.g. staff to child ratio, physical environment) and preschool teachers’ various pedagogical approaches, influence teachers’ choices to divide children into subgroups.

Materials and method

Sample

The current study used data from a Swedish survey conducted to examine preschool teachers’ perspectives in relation to group size in preschools (Williams, Sheridan, and Pramling Samuelsson 2016). The survey data was collected between 2012 and 2013. The sample for the current study consisted of 698 preschool teachers from 46 municipalities in Sweden who answered a web-based questionnaire (for descriptive information on the sample, see Nasiopoulou et al. 2017). The study follows the Swedish Research Council’s (2017) ethical regulations for humanities and social sciences.

The web-based questionnaire encompasses five themes: (1) preschool teachers’ background characteristics, (2) their working conditions, (3) their perspectives on organising children into subgroups (4) group sizes in their preschool and (5) the relationship between group size and children’s learning and development. For the purpose of the current study, one question was selected concerning preschool teachers’ considerations for organising children into subgroups: Do the following aspects play a role in group organising in preschool? For this question, 15 different aspects were given; these were selected to capture the complexity of preschool teachers’ everyday working context.

Analytical method

Hypothetical model

The aim of the present study was addressed by using CFA. CFA is an analytical method that is both theory driven and data driven. The researcher has, therefore, to specify a hypothetical model with predetermined factor structure according to the existing theory and previous research findings. CFA is then used to verify the number of underlying dimensions of latent factors and the pattern of variable latent factor relationships, such as factor loadings (Brown 2015).

This study is grounded in interactionist perspectives (Bergman, Magnusson, and El Khouri 2003), in which the individual acts and develops in dynamic interaction with the environment. Bronfenbrenner’s (1979, 1986) ecological systems theory lays the foundation for the study’s theoretical framework to understand the complexity of preschool teachers’ everyday practices and thereby multiple factors that may affect their decisions on classroom group arrangements. Using Bronfenbrenner’s (1979, 1986) theory, preschool teachers’ practices were considered, in this study, as embedded and shaped in a reciprocal interaction with multiple interrelated systems: macro-, exo-, meso-, micro- and chrono-. This implies that preschool teachers’ decisions to organise children into subgroups are closely linked to all levels of the entire early childhood education and care system – the individual and team level, the institutional level and the wider policy level – on preschools’ overall goals (Urban and Dalli 2012).

In addition to theoretical support, the study also drew from empirical studies (Pramling Samuelsson et al. 2016; Sheridan, Williams, and Pramling Samuelsson 2014; Williams, Sheridan, and Pramling Samuelsson 2018) examining preschool teachers’ perspectives on how they organise children into groups. These studies found that various aspects in the preschool context can affect preschool teachers’ daily practices in relation to their organising decisions. One aspect encompasses the child group composition, including children’s ages, gender, interests and needs. Other aspects include preschool teachers’ working conditions as schedules, administrative tasks and cooperation with colleagues, preschools’ organisational environments including staff to child ratio, the group size and preschools’ physical room arrangement. These studies, however, have only investigated a small sample of preschool teachers and were based solely on qualitative interviews. Thus, there is a need to further investigate the dimensionality of group organising as a practice in preschools. The current study used a nationally representative sample of preschool teachers in order to gain a deeper knowledge on this particular practice.

Based on these theoretical standpoints and empirical findings a hypothesised single-factor model was tested in order to investigate its fit for the current study’s 15 observed variables (i.e. factor indicators). As a data reduction method, CFA takes into account the interrelationships among observed variables by identifying a small number of latent variables or factors (i.e. underlying theoretical constructs; see Brown 2015). In other words, CFA captures the factor structure underlying a set of indicators. In CFA, the factor structure is assumed to be the same for all individuals.

The indicators used in the current study were selected in order to represent the complexity of preschool teachers’ everyday practice when organising children into subgroups. Thus, the 15 indicators of group organising were related to (1) children characteristics including age, gender, ability and interest, (2) preschool teachers’ interpersonal relationships measured by the indicators relation with children, relation with parents, relation between preschool teachers and relation with the head of the preschool, (3) preschool’s structural aspects and physical ecology measured by the indicators working environment, buildings, sound level and staff to child ratio and (4) preschool teachers’ workplace demands measured by time for each individual child, administrative tasks and possibilities to influence these tasks. Each indicator was measured using a four-point Likert scale where 1 indicates a strongly negative attitude and 4 a strongly positive attitude to each indicator. Table 1 shows the descriptive statistics for these indicators.

Table 1. Descriptive statistics for factor indicators in preschool group organising.

Factor Indicators^a	Mean	Standard Deviation
Children’s ages (GOage)	3.27	0.8
Children’s genders (GOsex)	1.97	0.9
Children’s interests (GOintr)	2.46	1
Children’s abilities (GOabil)	2.23	0.9
Teacher’s relation with head of preschool department (GOrebos)	2.16	1.1
Teacher’s relation with parents (GOrepar)	2.36	1.1
Teacher’s relation with children (GOrekid)	2.94	1.1
Teacher’s relation with colleagues (GOrete)	2.91	1.1
Working environment (GOwen)	3.44	0.8
Building (GObuil)	3.41	0.8
Sound level (GOsound)	3.38	0.9
Teacher/child ratio (GOratio)	3.68	0.6
Time for individual child (GOtid)	3.40	0.8
Administrative tasks (GOtask)	3.32	0.8
Possibilities to influence administrative tasks (GOposs)	3.24	0.9

Note: N = 698. ^aIndicators are Likert scale variables ranged from 1: strongly disagree to 4: strongly agree.

Model evaluation

The hypothetical model was statistically tested against available data by means of a set of model fit indices. The chi-square test with degree of freedom (χ²/df) is the traditional measure for evaluating the model fit. A statistically significant χ² indicates that the hypothetical model does not fit the data well. However, given the sensitivity of the χ² statistic, a combination of model fit indices with cut-off values are recommended for a model fit evaluation with a large sample size (Hooper, Coughlan, and Mullen 2008). The most frequently reported indices are the root mean square error of approximation (RMSEA), the standardised root mean square residual (SRMR) and comparative fit indices using the comparative fit index (CFI) and the Tucker-Lewis index (TLI; Brown 2015; Hooper, Coughlan, and Mullen 2008; Hox and Bechger 1998). Typically, an RMSEA cut-off value close to .60 or below, CFI and TLI values close to .95 or greater and SRMR values close to .08 or below indicate a reasonably good fit between the hypothetical model and observed data (Brown 2015; Hu and Bentler 1999). If the hypothetical model does not fit the data well, it can be revised and the latent structure of the model can be modified based on prior research by examining the standardised residuals and modification indices for each parameter (Brown 2015). In this process, fixed parameters associated with larger than a χ² critical value of 3.84 can be freely estimated in order to reduce the model’s χ² statistic until an improved model fit is reached (Hox and Bechger 1998).

Analytical process

Applying CFA as the analytical method, the factor structure of preschool teachers’ considerations for group organising was investigated by estimating a confirmatory factor model. CFA was performed using Mplus version 7.4 statistical package. All missing data were treated by default in Mplus through a full information maximum likelihood modelling technique (Muthén and Muthén 1998–2017).

A unidimensional CFA model with a single factor (General) underlying the study’s 15 indicators was specified first. The model was identified using the marker variable method by fixing the factor loading of each indicator for the latent factor to 1.0. This is the default method of model identification in Mplus. Evaluating the multiple fit indices for a single-factor model, the χ²/df ratio, SRMR, RMSEA, CFI and TLI values were not within the recommended critical values by the research literature. The model fit evaluation also showed large modification indices for the indicators related to preschool teachers’ interpersonal relationships and preschool’s structural aspects. These findings demonstrated that the fit of the model could be improved by freely estimating the respective parameters.

Based on study’s theoretical and empirical rationale, the model was modified through several steps. A new model was specified against the one-factor solution by adding a new factor (Interrelational) based on modification indices for indicators concerning preschool teachers’ interpersonal relationships. The re-specified model with two factors showed a decreased χ² statistic, df and improved fit indices, but was still outside the accepted thresholds. The two-factor model was then re-specified by adding a third factor (Structural), taking into account the standardised residuals and modification indices for indicators related to preschools’ structural aspects and physical ecologies. Table 2 presents the model fit information criteria associated with each of the CFA models indicated by the nested model structure.

Table 2. Fit indices for CFA models.

CFA models	χ2	df	RMSEA	CFI	TLI	SRMR
CFA model with one general factor	1253.406	89	0.137	0.644	0.580	0.097
CFA model with one general factor and one residual factor	675.443	86	0.099	0.820	0.780	0.072
CFA model with one general factor and two residual factors	246.672	81	0.054	0.949	0.934	0.039

Note: χ²:chi-square statistic, df: degree of freedom, RMSEA: root mean square error of approximation, CFI: comparative fit index, TLI: Tucker-Lewis index, SRMR: standardised root mean square residual.

Results

CFA was conducted in several steps and the initial hypothesised one-factor model was modified according to overall goodness-of-fit indices, standardised residuals and modification indices. The modified model revealed a three-factor solution with one General factor (parent model) and two residual factors (nested models) identified after taking into account the General factor (Table 2). The first residual factor was labelled Interrelational and is indicated by residuals of the variables related to preschool teachers’ interpersonal relationships, such as with children, other teachers, the preschool head and parents. The other residual factor was labelled Structural and is indicated by residuals of the variables related to preschool structural characteristics, such as the physical environment (buildings, sound), working environment, and teacher to child ratio. The three-factor model with these two residual factors contains a subset of freed parameters (indicators) of the parent model. The fit indices for the three-factor model indicated a reasonably good fit between the hypothetical model and the observed data.

Figure 1 shows the path diagram of the three-factor CFA model with standardised parameter estimates.

Figure 1. Path diagram with standardised parameter estimates of the three-factor model.

As is shown in the diagram, most of the indicators are strongly related to the General factor, except for GOintr and GOabil. They have low relation to the general factor, at .07 and .09, respectively. It should be noted that the residuals of these two indicators are correlated, which implies that they may share some common variance that can form a narrow dimension of the grouping considerations. A model without these two variables was also tested and the latent factor structure did not alter.

GOage and GOsex also showed relatively low relation to the General factor. Conventionally, a level of .30 is considered the cut-off point for the magnitude of standardised factor loadings in order to be characterised as an acceptable indicator for the underlying latent factor (Brown 2015). However, these factors were kept in the model according to previous research findings on group organising. The General factor explained 72% variance in GOtask and 5% variance in GOintr. On average, both the residual factors Structural and Interrelational explained 40% of the variances in their indicators. Thus, the model results convey two distinct dimensions of preschool teachers’ decisions on group organising: the preschool’s organisational environment (Structural factor) and the preschool teachers’ interpersonal relationships (Interrelational factor).

Discussion

Applying the CFA analytical approach, the purpose of this study was to identify latent factors that determine Swedish preschool teachers’ considerations when organising children into subgroups. The CFA found a three-factor solution with two residual factors (Interrelational and Structural) nested under the General factor. The residual factors were indicated by the residual variances in the indicators that remained after taking into account the General factor. Even though it seems that the General factor and the narrow residual factors have the same indicators, the parts of the variance explained by the three factors are completely different. This means that the two residual factors add valuable details to the model structure without changing the main hypothesis: that there is a single dimension underlying this study’s indicators. The model results conveyed two important additional aspects for consideration when organising groups in preschools, which are related to preschools’ organisational environments and preschool teachers’ interpersonal relationships. The results support the complexity of organising children into subgroups partly as a collectively negotiated practice (Urban and Dalli 2012) and partly as closely connected to the complexities of preschool classrooms’ actual conditions.

The Interrelational factor indicates that preschool teachers’ interpersonal relationships are distinct aspects in their everyday practice and may potentially have a powerful impact on how the organisation of children into groups is conducted. Considering preschool teachers’ relations with children, this finding is in line with previous research (Sheridan, Williams, and Pramling Samuelsson 2014; Williams, Sheridan, and Pramling Samuelsson 2018), which found that dividing children into small groups is often a shared issue communicated between the preschool teacher and the children. Working in small groups, as a practice negotiated with children, can also be seen as a way of managing and practicing children’s social interactions (Salminen et al. 2014), enhancing a secure climate (Alvestad et al. 2014) and supporting children’s engagement (Fuligni et al. 2012; Vitiello et al. 2012).

Additionally, this factor indicates that preschool teachers’ relationships with parents are an important aspect when preschool teachers structure the everyday preschool context into groups. The national preschool curriculum (Swedish National Agency for Education 2010, 2016) emphasises families’ involvement in preschool and the importance of providing parents with opportunities to exercise influence over how goals can be made concrete when planning pedagogical activities. In agreement with previous research (Alasuutari 2010), this finding could highlight a partnership principle in preschool teachers’ planning and organising of the everyday preschool context by recognising parents’ competence in creating conditions for children’s well-being, learning and development.

Furthermore, this factor indicating the relational dimension of group organising in preschool, points out the importance of team collaboration suggesting that preschool teachers’ decisions are the result of everyday negotiations, shared thinking and questioning over and in actual situations. Likewise, organising children into subgroups, as a practice influenced by preschool teachers’ relationships with the preschool head, can be understood in terms of confidence and perception of support provided by the head of the preschool in order to facilitate and encourage preschool teachers in their daily tasks. In line with Heikka, Halttunen, and Waniganayake (2016), this can include enhancing teamwork and clarifying responsibility areas for preschool staff; guiding curriculum work and organisational conditions in preschool, such as work shifts, routines and paid time for planning; as well as providing preschool staff with opportunities to continuing professional development.

The Structural factor shows the structural aspects and physical ecology of preschool and incorporates indicators related to preschool teachers’ working environment, the preschool building, sound level and adult to child ratio as important aspects when organising the children into small groups. This result is in accordance with previous studies (Alvestad et al. 2014; Sheridan, Williams, and Pramling Samuelsson 2014; Williams, Sheridan, and Pramling Samuelsson 2018), which have indicated that an interconnection between various structural components of preschool teachers’ work environments creates a workplace framework that either enables or hinders the practice of organising children into subgroups. According to these studies, these components include the number of children in the whole group, the group composition, staff stability, availability of teacher assistants when various disruptions occur in daily schedule and routines and preschools’ physical ecologies in terms of space, building design and architecture. Jonsdottir et al. (2015), further investigating the relationship between preschool pedagogical practices and noise level, found that a low level of noise had a higher association with the structure and organisation of activity types than the overall group size. Altogether, these aspects can be understood within the broad concept of the well-being of both preschool teachers and children. In supporting this consideration, previous studies on preschool teachers’ well-being indicated a positive association of their job-related satisfaction with stronger professional engagement and performance in the classroom (Dennis and O’Connor 2013).

The General factor further indicates the workplace demands, including the indicators administrative tasks, possibilities to influence these tasks and the time for each individual child, as additional influential aspects in preschool teachers’ practices when organising groups. Recognising the complexity of early childhood practice, this factor supports two additional but divergent aspects that create both boundaries and opportunities when organising children into groups. The first aspect references preschool teachers’ perceptions on their participation in decision-making processes facilitating their everyday working context. This everyday working context comprised of multiple supplementary tasks not included in the time spent with children, such as activity planning, reflective meetings or managing daily routines. The growing policy demand for documentation both as a task itself and as a means for evaluation and development of preschool quality requires additional workplace responsibilities and creates new layers of complexity to preschool teachers’ working context (Grant et al. 2018). The completion of these tasks can be time consuming (Pramling Samuelsson et al. 2016) and requires adequate non-contact time with children. In Swedish preschools, there is no uniformly regulated non-contact time with children, and the completion of these tasks is negotiated locally between staff and the preschool head in accordance with everyday practical needs. A shortage of preschool staff, for example, can make it difficult to organise the children into groups, and instead teachers may primarily work with the whole group and allow children to make their own peer groupings (Williams, Sheridan, and Pramling Samuelsson 2018).

The second aspect can be seen in relation to preschool teachers’ perceptions on time pressures with respect to the implementation of curriculum. The revised preschool curriculum places strong emphasis on documentation for preschool teachers to follow children’s development and learning. This implies that organising children into subgroups may be a pedagogical practice that gives preschool teachers the time to meet each individual child’s needs and interests in order to observe, follow and support each child’s learning and development (Alvestad et al. 2014; Williams, Sheridan, and Pramling Samuelsson 2018).

However, considering the General factor, it is noteworthy that the factor loadings for the indicators related to children’s characteristics (age, gender, ability and interest) were low, suggesting that these indicators have no salient relationship to the conjectured factor. According to Brown (2015), dropping the indicators with nonsignificant or nonsalient loading can improve the model’s fit. However, previous studies found that group composition is an important aspect in preschool teachers’ approaches to group work in order to enhance coherence (Salminen et al. 2014), harmony (Pramling Samuelsson et al. 2016), well-being and functioning (Aarts, Burk, and Riksen-Walraven 2016). This was the rationale for not removing these indicators from the analysis. Therefore, without underestimating the synergy of these indicators in preschool teachers’ considerations for group arrangements, this finding may suggest that these indicators are bundled within the other indicators in the current study, such as preschool structural aspects, or may be interlinked with preschool teachers’ pedagogical intentions. For instance, in preschools with low adult/child ratios; large, mixed-age groups; or a physical environment and design not on par with the number of children, organising children into age-homogeneous subgroups or with a balance between older children and toddlers has been used as an organisational and pedagogical approach to support children’s well-being and development (Pramling Samuelsson et al. 2016; Williams, Sheridan, and Pramling Samuelsson 2018).

In summary, the results show preschool teachers’ decisions to organise children into subgroups as closely related to both the organisational preschool environment and the importance to establish a relationship-focused workplace culture. The organisational workplace framework encompassed by decisions in the exo-system (Bronfenbrenner 1979, 1986) can create either opportunities or boundaries for this specific pedagogical practice. The relational dimension suggests that the grouping practice evolves as an outcome of a permanent attention on challenges in their immediate setting and awareness of pedagogical possibilities when they plan and organise children’s time in preschool (Sheridan et al. 2011).

Conclusions

This study adds to the findings from recent but limited studies examining preschool teachers’ perspectives on how they structure their everyday work using various grouping practices. Much of the existing literature used analytical approaches within the qualitative paradigm. In contrast, this study explored a large sample of preschool teachers’ considerations to support a comprehensive perspective on their practices when organising children into subgroups by underscoring the contribution of a factor analytical approach as a complementary analytical approach for future educational research.

Nevertheless, there are certain limitations to this study that need to be considered. First, the current study’s purpose was not the primary objective stated in the web-based questionnaire from which the data were drawn. Instead, the concept of organising children into groups was a subtheme of the questionnaire, which was conducted to explore preschool teachers’ perspectives on group sizes in preschools. Therefore, the indicators were not exhaustive of everything that might be used to measure the study’s purpose. Second, because there is a lack of studies involving factor analytical approaches, this study was primarily exploratory. Thus, conclusions should be made with caution, and replication is certainly needed. However, as the research supports the benefits of working in small subgroups in preschool, this study can be considered an initial step in conducting observational instruments to measure and evaluate group organising in preschools.

Using Bronfenbrenner’s (1979, 1986) ecological systems terminology, young children’s experiences in preschool are not determined solely by their immediate microsystems, but also by the larger community and society. From this perspective, the results of this study can be useful for enhancing quality in pre- and in-service early childhood education programmes by preparing and training preschool teachers to develop effective classroom organisation strategies and successfully actualise their knowledge into practice. This is important as the Swedish preschool curriculum sets out guidelines for preschool practices but does not provide a methodology on how preschool teachers can effectively accomplish these practices.

Disclosure statement

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

Additional information

Funding

This work was supported by the Swedish Research Council [grant number 2011-05535].