Organizational capacity supporting STEAM implementation in early childhood education: a mixed-methods study

1 Introduction

Over the past decade, STEAM (Science, Technology, Engineering, Arts, and Mathematics) education has become one of the key priorities in international education policy, closely associated with the development of twenty-first-century competencies and the quality of early childhood education. International organizations emphasize that cognitive development, creativity, critical thinking, and problem-solving skills fostered during early childhood provide the foundation for lifelong learning. Consequently, STEAM principles are increasingly being integrated into early childhood education curricula (National Research Council, 2012; OECD, 2025a, 2025b; UNESCO, 2021).

Despite the growing importance of STEAM education, its implementation remains uneven across educational institutions. Most existing studies have focused primarily on teachers’ competencies, children’s learning outcomes, or the effectiveness of individual STEAM interventions (Margot and Kettler, 2019; Shernoff et al., 2017). However, recent research suggests that STEAM implementation is shaped not only by teachers’ preparedness and the provision of material resources but also by broader organizational conditions, including leadership, professional learning, collaboration, and resource management (Deák and Kumar, 2024; Thibaut et al., 2018; Walshe et al., 2025).

Although organizational aspects of STEAM implementation have received increasing scholarly attention, there remains a lack of research examining the role of early childhood education leaders in the broader organizational conditions associated with STEAM implementation. Furthermore, there is a shortage of mixed-methods studies that integrate national-level quantitative data with qualitative insights from educational leaders to examine resource provision within the broader organizational context of STEAM implementation.

To address this research gap, the present study aims to examine the organizational conditions associated with STEAM implementation in early childhood education institutions by integrating quantitative data from a national survey with qualitative insights from educational leaders. The mixed-methods design combines national-level evidence on teaching and learning resource provision with educational leaders’ perspectives on the broader organizational conditions surrounding STEAM implementation. In this way, the study situates resource provision within a wider context that includes leadership practices, professional learning, collaboration, and other organizational conditions.

1.1 Research objectives

The aim of this study was to examine the provision of teaching and learning resources and to explore the broader organizational conditions associated with STEAM implementation in Lithuanian early childhood education institutions.

Accordingly, the study pursued three objectives:. This study contributes to the STEAM education literature by applying an organizational capacity perspective to the examination of STEAM implementation in early childhood education. Unlike most previous studies, which have examined organizational factors in isolation, this research considers organizational leadership, teachers’ professional learning, collaboration, human and material resources, learning environments, partnerships, and organizational and financial conditions as interconnected dimensions of the broader organizational context of STEAM implementation. Furthermore, the mixed-methods research design enables the integration of national-level quantitative evidence with qualitative insights from early childhood education leaders, providing a more comprehensive understanding of how resource provision can be situated within this broader organizational context.

• to describe national-level patterns in the provision of teaching and learning resources across Lithuanian early childhood education institutions, with more detailed analysis of the thematic area “Exploring and Understanding the Environment”;
• to explore educational leaders’ perspectives on the resources, leadership practices, professional learning, collaboration, and barriers associated with STEAM implementation;
• to integrate the quantitative and qualitative findings to situate resource provision within a broader organizational understanding of the conditions associated with STEAM implementation in early childhood education institutions.

2 Background

2.1 Organizational leadership and organizational capacity in the implementation of STEAM education

In recent years, educational research has increasingly adopted the concept of organizational capacity to understand the organizational conditions associated with the implementation of educational innovations. This theoretical perspective emphasizes that educational improvement is associated not only with individual resources or innovations but also with an organization's capacity to mobilize human, material, and organizational resources, develop a shared vision, foster professional learning, and manage change strategically (Harris, 2001; Leithwood et al., 2020; Mitchell and Sackney, 2011). Organizational capacity is understood as a dynamic organizational attribute associated with continuous learning and adaptation to evolving educational needs (Fullan, 2020a, 2020b). Consequently, contemporary education policy increasingly recognizes that educational improvement involves both resource provision and an organization's capacity to create learning-oriented environments, strengthen professional communities, and systematically develop organizational capacity (OECD, 2025a).

This organizational capacity perspective has also gained increasing prominence in STEAM education research. Recent systematic literature reviews indicate that STEAM implementation is associated with multiple organizational dimensions, including leadership, professional learning, a culture of collaboration, learning environments, and resource management (Deák and Kumar, 2024; Margot and Kettler, 2019; Walshe et al., 2025). From this perspective, STEAM implementation is understood not merely as the application of individual teaching methods or technologies but as an organizational process in which diverse organizational resources and conditions coexist and interact.

Teachers’ professional competencies are widely recognized as one of the core components of organizational capacity. However, recent studies suggest that STEAM implementation involves professional competence alongside broader professional and organizational conditions. Teachers’ ability to deliver integrated STEAM education is closely associated with professional self-efficacy, continuous professional learning, and organizational opportunities to reflect on practice, collaborate, and learn from colleagues (Margot and Kettler, 2019; Movahedazarhouligh et al., 2023; Shernoff et al., 2017; Wang and Lin, 2026). Consequently, professional learning is increasingly conceptualized as an organizational process associated with the development of an institution's collective capacity for educational innovation.

A similar shift can be observed in the understanding of material resources. Although specialized STEAM materials, laboratory equipment, and digital technologies can enrich teaching and learning, their availability represents only one dimension of the conditions associated with STEAM implementation. Research suggests that their educational value is closely related to how they are integrated into inquiry-based, problem-based, and collaborative learning experiences (Bybee, 2013; Deák and Kumar, 2024). In early childhood education, research also emphasizes the relevance of flexible learning environments, open-ended materials, and natural resources that encourage children's initiative, creativity, and experiential learning (DeJarnette, 2018; Walshe et al., 2025). These findings suggest that material resources can be understood as part of a broader pedagogical and organizational context rather than as an isolated dimension of STEAM implementation.

Overall, contemporary STEAM literature conceptualizes organizational capacity as encompassing leadership, professional learning, teachers’ competencies, a culture of collaboration, and strategic resource management. This perspective broadens attention from the availability of individual resources or technologies to include how organizations mobilize human, material, and organizational resources within the broader context of STEAM implementation. It is this organizational capacity perspective that provides the theoretical foundation for the present study's examination of the organizational conditions associated with STEAM implementation in early childhood education institutions.

2.2 Leadership as a dimension of organizational capacity for STEAM implementation

Leadership is widely recognized as an important dimension of organizational capacity and educational change. Contemporary theories of educational leadership emphasize that the role of educational leaders extends beyond administrative responsibilities to encompass setting a strategic direction, promoting professional learning, fostering a culture of collaboration, and creating conditions that support teachers’ professional growth (Fullan, 2020a; Hallinger, 2011; Leithwood et al., 2020). From this perspective, leadership can contribute to building organizational capacity by creating conditions that enable institutions to respond to evolving educational needs.

In the context of early childhood education, leadership is relevant to the organizational conditions within which teachers engage in daily practice, collaborative work, and professional dialogue. Leaders of early childhood education institutions can support the development of professional learning communities by bringing teachers together around a shared educational vision, encouraging reflective practice, and supporting continuous professional development (Thornton, 2023). Research indicates that professional learning communities in early childhood education are associated with leadership practices that promote distributed leadership, collegial learning, reflective practice, and teachers’ active participation in decision-making (Heikka and Suhonen, 2019; Spillane, 2006; Thornton, 2023).

This perspective aligns closely with the theory of professional learning communities, which conceptualizes professional learning as an organizational rather than an individual process. Professional learning communities emphasize the systematic examination of practice, reflection on teaching and learning processes, and the sharing of professional knowledge and experience (Stoll et al., 2006). Similarly, the concept of the learning organization emphasizes learning across different levels of the organization as an important component of educational change (Senge, 2006). Recent studies in early childhood education further highlight the relevance of leadership-supported professional learning communities to teachers’ engagement with educational innovations (Thornton, 2023).

Leadership is relevant to the organizational context of STEAM education because integrated teaching and learning involve not only subject-specific knowledge but also continuous professional learning, collaborative planning, and ongoing reflection. Recent systematic reviews highlight the relevance of organizational conditions supporting collegial collaboration, professional learning, and distributed leadership in STEAM implementation (Deák and Kumar, 2024; Margot and Kettler, 2019; Movahedazarhouligh et al., 2023; Shernoff et al., 2017). Empirical studies likewise associate leadership support for professional dialogue, collaborative planning, and teachers’ reflective practice with organizational conditions conducive to the integration of STEAM into educational practice (Deák and Kumar, 2024; Thibaut et al., 2018). Such organizational environments may also be associated with teachers’ professional self-efficacy and readiness to engage with inquiry-based and integrated approaches to teaching and learning (Bandura, 1997; OECD, 2025a; Tschannen-Moran and Hoy, 2001; Wang and Lin, 2026).

Overall, leadership in early childhood education can be understood as one dimension of organizational capacity associated with professional learning, a culture of collaboration, and strategic change management. Accordingly, leadership can be considered alongside teachers’ competencies, professional learning, collaboration, material resources, and other organizational conditions when examining STEAM implementation in early childhood education.

2.3 Contemporary perspectives on STEAM education and the organizational conditions for its implementation

Over the past decade, the concept of STEAM education in early childhood education has evolved considerably. Recent systematic reviews emphasize that STEAM is no longer understood merely as the integration of five disciplinary areas. Instead, it is increasingly conceptualized as a pedagogical approach grounded in inquiry, play, creativity, problem-solving, and experiential learning, enabling children to actively construct knowledge and apply it in authentic contexts (Johnston et al., 2022; Walshe, 2024; Walshe et al., 2025). This perspective aligns closely with the philosophy of early childhood education, in which learning is driven by children's initiative, curiosity, and active engagement with their environment. Although some of the literature cited in this study addresses integrated STEM rather than STEAM specifically, it is included where it examines organizational, pedagogical, or professional conditions relevant to interdisciplinary implementation. The present study nevertheless focuses specifically on STEAM, reflecting the explicit inclusion of the arts within the early childhood education context examined (English, 2016; Honey et al., 2014).

Alongside this conceptual shift, views on the resources required for STEAM education have also changed. Earlier research primarily emphasized specialized teaching materials and digital technologies; however, more recent studies indicate that technology represents one component of a broader set of pedagogical and material conditions associated with STEAM education. Increasing attention is also given to learning environments that enable children to experiment, design, manipulate diverse materials, engage in project-based activities, and collaborate with peers. Within this framework, digital technologies are regarded as pedagogical tools that can complement inquiry-based learning alongside other materials and learning environments (Deák and Kumar, 2024; Johnston et al., 2022). This perspective is particularly relevant in early childhood education, where open-ended materials, natural resources, and learning environments that stimulate children's creativity are particularly relevant.

Increasing attention has also been devoted to the social and organizational dimensions of STEAM education. Research indicates that STEAM implementation is associated not only with individual teachers’ competencies but also with broader organizational conditions, including professional learning communities, partnerships, and opportunities for continuous collaborative learning. In early childhood education, professional STEAM networks and collaboration with universities, museums, and other organizations are regarded as potential sources of professional learning, knowledge exchange, and support for STEAM-related practices (O'Neill et al., 2024).

Despite the growing body of research on STEAM education, its organizational dimension remains insufficiently explored. Most empirical studies focus on teachers’ competencies, children's learning outcomes, or individual STEAM interventions, whereas the broader organizational conditions associated with STEAM implementation receive considerably less attention (Margot and Kettler, 2019; Movahedazarhouligh et al., 2023; Shernoff et al., 2017). In particular, there is a shortage of research examining the role of early childhood education leaders in strategic resource planning, organizing professional learning, fostering collaborative cultures, and coordinating STEAM-related activities at the institutional level.

The literature also reveals an important methodological gap. Mixed-methods studies integrating national-level quantitative data with qualitative insights from educational leaders remain relatively rare. Such research designs can combine broader-scale quantitative evidence with qualitative insights into the organizational context in which STEAM implementation takes place.

In response to these theoretical and methodological gaps, the present study examines resource provision and the broader organizational conditions associated with STEAM implementation. This mixed-methods research design integrates national-level evidence on resource provision with educational leaders’ perspectives on leadership practices, professional learning, collaboration, learning environments, and other organizational conditions associated with STEAM implementation.

3 Materials and methods

3.1 Research design, context, and participants

The empirical study focused on education institutions implementing the national Early Childhood Education Curriculum in Lithuania. The study aimed to examine the organizational conditions supporting STEAM implementation by combining national-level evidence on the provision of teaching and learning resources with educational leaders’ perspectives on the broader organizational context in which these resources are selected, used, and developed.

The study employed a sequential mixed-methods design in which the quantitative phase preceded and provided the empirical context for the subsequent qualitative phase. The quantitative component was used to identify national-level patterns in the provision of teaching and learning resources across Lithuanian early childhood education institutions. Following the preliminary analysis of the quantitative data, the qualitative component was conducted to contextualize the national-level findings and to explore educational leaders’ perspectives on the broader organizational conditions surrounding STEAM implementation.

The two components addressed complementary dimensions of the research problem. The quantitative phase focused specifically on the provision of teaching and learning resources, providing a national-level description of the material conditions relevant to STEAM implementation. The qualitative phase extended the inquiry beyond resource provision by examining how educational leaders perceived and managed resources within a broader organizational context, including leadership practices, teachers’ professional learning, collaboration, motivation, strategic planning, and barriers to STEAM implementation. Thus, the qualitative component was not intended to measure the same constructs as the quantitative survey. Rather, it contextualized and extended the quantitative evidence by examining the organizational processes within which resource provision acquires practical significance. Accordingly, the quantitative component addressed the first research objective, the qualitative component addressed the second objective, and the integration of both components addressed the third objective.

The integration of the two components was based on complementarity and sequential interpretation. Quantitative findings provided the empirical starting point for identifying patterns of resource provision, whereas qualitative data were used to explore the organizational conditions surrounding the selection, use, and development of these resources at the institutional level. Integration therefore enabled the study to move from a national-level description of resource provision toward a broader interpretation of the organizational capacity supporting STEAM implementation in early childhood education institutions (Creswell and Creswell, 2023; Creswell and Poth, 2018; Patton, 2014; Žydžiūnaitė and Sabaliauskas, 2017).

The quantitative data were obtained from the Lithuanian National Agency for Education, which conducted a nationwide survey of educational institutions. The purpose of the survey was to assess institutional preparedness for implementing the Educational Provision Standard for Schools Implementing Early Childhood, Pre-primary, and General Education Programmes. This standard, approved by the Minister of Education, Science and Sport of the Republic of Lithuania on 4 December 2024 (Order No. V-1374), constitutes a national education policy framework defining requirements for educational provision.

Within the context of this national standard, the present study considered the provision of teaching and learning resources as one component of the broader organizational conditions relevant to curriculum and STEAM implementation.

Survey data were collected from institutional representatives responsible for educational management, resource planning, and curriculum implementation. The quantitative sample comprised 831 education institutions, providing a national dataset on the provision of teaching and learning resources. The dataset enabled the identification of broader patterns in resource provision among participating Lithuanian educational institutions implementing the Early Childhood Education Curriculum. The results of the national survey have been published by the Lithuanian National Agency for Education, ensuring transparency and public accessibility of the quantitative data source used in the present study. https://www.nsa.smsm.lt/wp-content/uploads/2025/07/Svietimo-aprupinimo-standarto-igyvendinimas-2025-m.pdf

The qualitative component involved 38 leaders of early childhood education institutions, all of whom represented institutions included in the national quantitative dataset of 831 participating institutions. An invitation to participate in the qualitative phase was distributed to a broader group of leaders of institutions included in the quantitative study, and participation was voluntary. A total of 38 leaders responded to the invitation and completed the qualitative survey. Thus, the qualitative sample constituted a voluntary response sample rather than a purposively selected or statistically representative subsample of the national dataset. This voluntary participation may have introduced self-selection bias, as leaders who chose to respond may have differed systematically from those who did not participate, for example in their level of interest in, engagement with, or experiences of STEAM implementation. Consequently, the qualitative findings should be interpreted as reflecting the perspectives of the participating leaders rather than as representative of all institutions included in the national dataset. Participants were not selected on the basis of their institutions’ level of resource provision; therefore, variation in resource provision was not used as a formal sampling criterion for the qualitative phase.

The participating leaders represented early childhood education institutions located in different geographical areas of Lithuania, providing perspectives from a range of institutional contexts. However, geographical diversity was not used as a formal sampling criterion. The qualitative component was intended to provide contextual and interpretive depth rather than statistical representativeness (Baškarada, 2014; Yin, 2018).

The leaders’ responses provided insight into how educational leaders perceived the resources required for STEAM implementation and the broader organizational conditions associated with their use, including leadership practices, professional learning, collaboration, teacher motivation, organizational barriers, and strategic improvement initiatives. In this way, the qualitative component extended the quantitative analysis of resource provision by situating material resources within the broader organizational context of STEAM implementation.

3.2 Research sample, duration, and procedure

The quantitative component of the study was conducted as a national survey administered from 21 October to 21 November 2025 through the Education Management Information System (EMIS). The survey instrument was based on the Educational Provision Standard approved in 2024 (Lietuvos Respublikos švietimo, mokslo ir sporto ministerija, 2024) and included the essential teaching and learning resources specified for the implementation of the Early Childhood Education Curriculum. The five thematic areas presented in Table 1 were not generated by the researchers as analytical categories but were derived directly from the structure of the national Early Childhood Education Curriculum and the corresponding Educational Provision Standard. Teaching and learning resources were grouped according to the thematic area to which they were assigned in these national policy documents. Educational institutions reported the quantity of resources currently available and their additional resource needs, taking into account the number of children, learning spaces, and the organization of educational activities. A total of 1,116 educational institutions implementing the Early Childhood Education Curriculum were requested to provide data. Of these, 831 institutions submitted data, while 285 institutions did not provide data. The resulting participation rate was 74.46%. The unit of analysis in the quantitative component was the early childhood education institution rather than an individual respondent. Each institution could submit only one institutional response; therefore, the quantitative sample comprised N = 831 institutional responses representing 831 educational institutions implementing the Early Childhood Education Curriculum. Following the quantitative phase, the qualitative component was conducted from 15 May to 1 June 2026. The qualitative data collection followed the preliminary analysis of the quantitative data and was intended to contextualize and extend the national-level findings by examining educational leaders’ perspectives on the broader organizational conditions surrounding STEAM implementation. The qualitative component therefore broadened the focus from the provision of teaching and learning resources to the ways in which resources were perceived and used within the organizational context of early childhood education institutions.

Table 1. National-level provision of teaching and learning resources across thematic areas.

“No mandatory resources” indicates that the Educational Provision Standard does not specify mandatory resources for individual children or small groups within the thematic area “Me and My Community”; therefore, no resource provision level was calculated.

The qualitative component involved 38 educational leaders of early childhood education institutions included in the quantitative dataset. Participation was voluntary following an invitation distributed to a broader group of leaders from institutions participating in the quantitative phase. All participants represented early childhood education institutions. Geographically, the participating institutions were located in 11 Lithuanian localities, including three major cities and eight smaller towns, thereby providing perspectives from different geographical and institutional contexts. Geographical location, however, was not used as a formal sampling criterion, and the qualitative sample was not intended to be statistically representative of Lithuanian early childhood education institutions. Information on institution size and participants’ years of leadership experience was not collected.

Each participating leader responded to ten open-ended questions (see Table 2). The questions explored leaders’ perspectives on resources considered important for STEAM implementation, perceived resource shortages, leadership practices, professional learning, collaboration, teacher motivation, barriers to implementation, and organizational improvement initiatives. The questions were developed by the authors based on the study objectives and relevant literature. The questionnaire was not formally piloted prior to data collection, and no separate content-validation procedure involving external experts was conducted. The open-ended format was intended to allow participants to describe their experiences and perspectives in their own terms within the predefined areas of inquiry.

Table 2. Survey questions for educational leaders.

The qualitative questions were informed by the literature on STEAM implementation, organizational capacity, professional learning, and educational leadership, as detailed in Table 2. The preliminary quantitative findings provided the empirical context for the subsequent qualitative inquiry rather than serving as a basis for selecting participants according to their institutions’ resource provision levels. Accordingly, the qualitative component was used to provide contextual and interpretive depth to the national quantitative findings and to examine the broader organizational conditions within which teaching and learning resources are mobilized for STEAM implementation.

Integration of the quantitative and qualitative components was based on complementarity and occurred primarily at the interpretation stage. The quantitative data were used to identify national-level patterns in the provision of teaching and learning resources, whereas the qualitative data extended these findings by providing educational leaders’ perspectives on the broader organizational conditions within which resources are selected, used, and developed. The two datasets were therefore not treated as equivalent measures of the same constructs. Instead, quantitative findings concerning resource provision were considered alongside qualitative categories related to material and human resources, learning environments, professional learning, collaboration, leadership practices, and perceived barriers to STEAM implementation. This approach enabled the quantitative findings to be situated within a broader organizational context and allowed areas of convergence, complementarity, and divergence between the two components to be considered during interpretation (Creswell and Creswell, 2023; Creswell and Poth, 2018).

3.3 Data analysis

Quantitative data were analyzed descriptively using the dataset provided by the National Agency for Education. For each teaching and learning resource, the resource provision level was defined as the percentage of the identified resource need that was met within an institution. It was calculated by dividing the quantity of the resource available in an institution by the quantity identified as required for that institution and multiplying the result by 100. Thus, the percentage represents the extent to which the identified resource need was met rather than the percentage of institutions possessing the resource. For example, if an institution had one unit of a particular resource but required five units, its resource provision level was calculated as (1/5) × 100 = 20%. The national-level percentage for each resource was calculated by the National Agency for Education as the arithmetic mean of the institution-level provision percentages across the participating institutions. No inferential statistical analyses were performed in the present study.

Qualitative data were analyzed using a hybrid deductive–inductive qualitative content analysis. The deductive component was provided by the broader thematic focus of the study and the literature-informed open-ended questions, which addressed resources, leadership practices, professional learning, collaboration, teacher motivation, barriers, and organizational improvement related to STEAM implementation. Within these broader areas, however, codes and subcategories were not predetermined but were developed inductively from participants’ responses.

The unit of analysis was a meaning unit, defined as a segment of a participant's response expressing a distinct idea relevant to the research objectives. The responses were organized and coded using Microsoft Excel. Both authors participated collaboratively in the coding process. Rather than coding the full dataset independently, the authors jointly reviewed the responses, discussed the interpretation of meaning units, and refined the coding structure through an iterative process. Similar meaning units were assigned descriptive codes, and codes reflecting related meanings were subsequently grouped into subcategories. Conceptually related subcategories were then organized into broader categories representing the main patterns identified in the data. The coding structure was reviewed and refined collaboratively by both authors to ensure that the categories remained grounded in participants’ responses while also addressing the broader areas of inquiry defined by the study. Coding decisions and disagreements in interpretation were discussed collaboratively until agreement was reached, and the coding structure was iteratively reviewed against the original responses. However, the authors did not independently code the full dataset; therefore, formal inter-coder reliability statistics, such as Cohen's kappa, were not calculated. Member-checking with participants was also not conducted.

For the presentation of qualitative findings, frequencies were calculated at the participant level. The value n therefore indicates the number of participants whose responses contained at least one meaning unit coded within a given subcategory, rather than the number of individual statements or coded segments. Because a participant could refer to more than one issue, the same participant could be represented in multiple subcategories. Consequently, the reported frequencies are not mutually exclusive and were used to indicate the prevalence of themes within the qualitative sample rather than to rank their relative importance. Illustrative quotations were selected to demonstrate the meaning and content of the identified subcategories.

3.4 Research ethics

The study was conducted in accordance with the research ethics principles established by Vytautas Magnus University and received ethical approval from the Ethics Committee of the Institute of Educational Research (Approval No. SA-EK-26-96, 7 May 2026). Throughout all stages of the study, the principles of confidentiality, data protection, and voluntary participation, and respect for participants’ rights were observed.

The quantitative data were obtained from the Lithuanian National Agency for Education, which conducted the nationwide survey of educational institutions. The dataset used in the present study contained no personally identifiable information. The quantitative data were analyzed solely for the purposes of the present research and were handled in accordance with applicable data protection requirements.

Participation in the qualitative component of the study was voluntary. Before data collection, the leaders of early childhood education institutions were informed about the purpose of the study, the nature of their participation, the procedures for data collection and use, and their right to withdraw from the study without adverse consequences. Written informed consent was obtained from all participants prior to participation. The qualitative responses were treated confidentially and analyzed in a manner that did not disclose the identities of individual participants or their institutions.

4 Results

The Results section first presents national-level descriptive findings on the provision of teaching and learning resources across the thematic areas of the Early Childhood Education Curriculum. This overview is used to identify variation in resource provision across curriculum areas.

Following this overview, a more detailed descriptive analysis is presented for the thematic area “Exploring and Understanding the Environment.” This area was selected for further examination after reviewing the national survey results. The selection was based on two considerations. First, the thematic area is closely related to inquiry, exploration, experimentation, and problem-solving activities that are particularly relevant to STEAM-oriented educational practice. Second, the quantitative results showed neither uniformly high nor uniformly low levels of resource provision in this area, allowing a more detailed examination of both available resources and reported additional resource needs. The selection of this thematic area was therefore analytical and illustrative rather than intended to imply that it was uniquely representative of all curriculum areas.

The data presented in Table 1 illustrate the provision of two types of teaching and learning resources across the five thematic areas of the Early Childhood Education Curriculum: demonstration resources and resources intended for individual children or small groups.

For demonstration resources, the highest reported level of resource provision was observed in the thematic area “Creating and Expressing” (85.73%). Lower and relatively similar provision levels were reported for “Exploring and Understanding the Environment” (66.24%), “Our Health and Well-being” (65.18%), and “Me and My Community” (65.10%). The lowest level of provision for demonstration resources was reported for “Me in the World of Languages” (58.37%).

For resources intended for individual children or small groups, the highest reported level of resource provision was observed in “Creating and Expressing” (82.31%), followed by “Me in the World of Languages” (78.78%), “Our Health and Well-being” (70.92%), and “Exploring and Understanding the Environment” (67.52%). For the thematic area “Me and My Community,” the Educational Provision Standard does not specify mandatory resources in this resource category; therefore, no percentage is reported in Table 1.

Overall, the descriptive results indicate variation in the reported levels of teaching and learning resource provision across thematic areas and resource types. “Creating and Expressing” showed the highest reported provision levels for both resource categories, whereas the lowest reported provision level for demonstration resources was observed for “Me in the World of Languages.” These findings are descriptive and do not, by themselves, establish the pedagogical or organizational reasons underlying the observed differences.

The following section presents a more detailed descriptive analysis of resource provision within the thematic area “Exploring and Understanding the Environment.”

The data presented in Table 3 show variation in the reported levels of provision of specific teaching and learning resources within the thematic area “Exploring and Understanding the Environment.”

Table 3. Provision levels of teaching and learning resources for the thematic area “exploring and understanding the environment” in early childhood education institutions (%).

Among demonstration resources, the highest reported provision level was observed for posters or cards supporting number learning (77.24%), followed by interactive posters, maps, picture cards, labels, or cubes for exploring natural and human-made objects (70.96%). Nature calendars had a provision level of 60.20%, while prism sets for exploring the spectrum of light and colours showed the lowest provision level among demonstration resources (44.37%).

Among resources intended for individual children or small groups, several types showed provision levels above 70%. These included resource kits for identifying domestic and wild animals (73.66%), vegetables and fruits (73.03%), conducting nature investigations (72.76%), exploring the characteristics of objects in the local environment (72.43%), and number puzzles (70.41%). Resource kits for identifying trees, shrubs, and medicinal plants had a provision level of 58.12%.

The lowest reported provision level among all resources presented in Table 3 was observed for STEAM laboratory kits designed for exploring natural and physical phenomena, such as force, solar and water energy, and simple machines. The provision level for these kits was 30.01%. This indicates that, on average, the reported quantity of STEAM laboratory kits met approximately 30% of the identified need across the participating institutions. This percentage indicates that specialized STEAM laboratory kits had a lower provision level than the other teaching and learning resources included in this thematic area. However, the provision level of a specific type of resource should not be interpreted as a direct measure of the extent or quality of STEAM implementation within an institution.

Overall, the descriptive findings demonstrate substantial variation in the provision levels of specific resources within the thematic area “Exploring and Understanding the Environment.” While several resource types showed provision levels above 70%, other resources, particularly STEAM laboratory kits and prism sets, showed lower provision levels. The pedagogical and organizational implications of these differences are considered in the Discussion section.

The qualitative content analysis identified five overarching categories of resources perceived by educational leaders as relevant to STEAM implementation: human resources, material resources, learning environments, financial and organizational resources, and partnerships (Table 4). Within these categories, leaders referred to several distinct types of resources, indicating that their accounts of STEAM implementation extended beyond material provision alone.

Table 4. Early childhood education leaders’ perspectives on the resources supporting the implementation of STEAM education (n = 38).

n indicates the number of participants whose responses were coded within each subcategory. Because individual responses could contain references to more than one type of resource, a participant's response could be coded into multiple subcategories; therefore, frequencies across subcategories are not mutually exclusive and should not be interpreted as a ranking of importance.

The most frequently identified subcategory was STEAM resources and digital technologies (n = 27). Leaders referred to educational robots, LEGO Education kits, microscopes, tablets, interactive boards, educational software, and other digital or scientific resources available in their institutions. Teachers’ STEAM competencies and understanding of the STEAM approach constituted the second most frequently identified subcategory (n = 23). Leaders also referred to STEAM learning spaces and inquiry-based learning environments (n = 17), open-ended, natural, and everyday materials (n = 16), teachers’ creativity, initiative, and community engagement (n = 14), collaboration with external partners (n = 12), and funding and project-based initiatives (n = 10).

Although material and technological resources constituted the most frequently identified subcategory, leaders’ responses also highlighted human resources. In particular, respondents described teachers’ understanding of the STEAM approach, professional competencies, creativity, initiative, and willingness to organize STEAM activities as relevant to implementation. These responses indicate that leaders perceived material resources and teachers’ professional capacities as distinct but co-occurring dimensions of the organizational conditions surrounding STEAM implementation.

Leaders also identified the learning environment as relevant to STEAM implementation. Responses referred to dedicated STEAM rooms or laboratories, indoor and outdoor learning spaces, and inquiry-oriented environments. In addition, some leaders emphasized the use of open-ended, natural, and everyday materials, including materials available in the natural environment and everyday objects.

Financial and organizational resources and external partnerships were identified less frequently. Leaders referred to funding and project-based initiatives (n = 10) and collaboration with external partners (n = 12), including universities, museums, families, and other organizations.

Overall, the qualitative findings show that educational leaders identified multiple types of resources in relation to STEAM implementation. Material and technological resources constituted the most frequently identified subcategory, while human resources, learning environments, financial and organizational resources, and partnerships were also represented in their responses. These findings provide a broader organizational context for interpreting quantitative patterns of resource provision in relation to STEAM implementation.

The qualitative content analysis identified five overarching categories of leadership practices described by early childhood education leaders in relation to STEAM implementation: strategic leadership, professional learning, collegial collaboration, teacher motivation, and change management (Table 5).

Table 5. Leadership practices reported by early childhood education leaders in relation to STEAM implementation (n = 38).

The most frequently identified subcategory was sharing good practice and mentoring within the category of collegial collaboration (n = 21). Leaders described practices such as peer-to-peer learning, joint planning, mentoring, peer observation, open lessons, and the sharing of educational experiences. These responses indicate that collegial collaboration was a prominent feature of leaders’ accounts of organizational support for STEAM implementation.

Systematic professional development was identified in the responses of 20 participants. Leaders referred to seminars, training programmes, conferences, online courses, and other forms of professional development related to STEAM education. Learning through practice and reflection was identified in the responses of 16 participants and included institution-based professional learning, professional discussions, the practical application of new approaches, and subsequent reflection.

Strengthening the STEAM culture within the organization, categorized under change management, was identified in the responses of 18 participants. Leaders described efforts to integrate STEAM-related activities into everyday educational practice and to develop a more institution-wide approach rather than relying on isolated or fragmented activities.

Leadership practices related to teacher motivation were represented by two subcategories. Fostering a culture of trust and support was identified in the responses of 17 participants. Leaders referred to supporting teachers’ ideas and initiatives, demonstrating trust in their professional practice, and recognizing their efforts. Creating conditions for professional growth was identified in the responses of 15 participants and included providing professional development opportunities, encouraging collaboration, and allocating resources and time for collaborative activities.

Strategic leadership was identified in the responses of 13 participants. Leaders referred to the inclusion of STEAM principles and activities in institutional strategies, annual action plans, curriculum documents, and professional development planning.

Overall, the qualitative findings show that leaders described multiple organizational practices in relation to supporting STEAM implementation. Sharing good practice and mentoring, systematic professional development, and strengthening the STEAM culture within the organization were among the most frequently identified subcategories. Practices related to teacher motivation, professional learning through practice and reflection, and strategic leadership were also represented in participants’ responses. Taken together, these findings illustrate the range of organizational practices that educational leaders associated with STEAM implementation.

As shown in Table 6, the qualitative content analysis identified several categories of barriers described by educational leaders in relation to STEAM implementation, encompassing material, pedagogical, organizational, and financial factors. The most frequently identified subcategory was the limited provision of specialized STEAM resources and digital technologies (n = 17), followed by the lack of dedicated STEAM learning spaces and laboratory equipment (n = 15). Limited teachers’ STEAM competencies and understanding of the STEAM approach were identified in the responses of 12 participants.

Table 6. Early childhood education leaders’ perspectives on the barriers to the implementation of STEAM education (n = 38).

Organizational and financial barriers were also represented in leaders’ responses. Insufficient time for planning and collaboration was identified in the responses of nine participants, while insufficient funding was identified in the responses of six participants. Leaders referred to limited time for preparing STEAM activities and engaging in collaborative professional learning, as well as financial constraints affecting the acquisition of specialized materials and equipment.

A smaller group of participants (n = 4) reported no significant barriers or indicated that material resource limitations were not perceived as the primary barrier in their institutional context. These responses indicate variation in how participating leaders perceived material constraints. Given the small number of participants represented in this subcategory, these responses should be interpreted as reflecting a limited set of perspectives within the qualitative sample rather than as a general pattern.

Overall, the qualitative findings indicate that leaders described multiple and potentially co-occurring barriers to STEAM implementation. Material resource constraints constituted the most frequently identified barriers, particularly those related to specialized STEAM resources and digital technologies and to dedicated learning spaces and laboratory equipment. At the same time, pedagogical, organizational, and financial barriers were also represented in participants’ responses. The findings therefore indicate that educational leaders perceived barriers to STEAM implementation across several dimensions rather than as a single type of constraint.

Taken together, the findings presented in Tables 4–6 identify several dimensions that educational leaders associated with STEAM implementation, including material and human resources, learning environments, professional learning, collaboration, leadership practices, and perceived barriers. To visually summarize the main dimensions identified across the qualitative findings, Figure 1 presents a data-informed conceptual framework. The framework is intended as an interpretive synthesis of the findings rather than as a tested causal or explanatory model.

Figure 1 presents a data-informed conceptual framework synthesizing the main organizational dimensions identified across the study findings. The framework brings together leadership practices, human and material resources, learning environments, professional learning and collaboration, and perceived barriers associated with STEAM implementation in early childhood education institutions. Rather than assigning causal or hierarchical relationships among these dimensions, the framework illustrates how they were represented as interconnected aspects of the organizational context surrounding STEAM implementation.

Leadership-related practices included strategic planning, support for professional learning, collegial collaboration, teacher motivation, and efforts to strengthen STEAM within the organizational culture. Human-resource dimensions included teachers’ STEAM competencies, understanding of the STEAM approach, creativity, and initiative, while material dimensions included specialized STEAM resources, digital technologies, and learning environments. The findings also identified material, pedagogical, organizational, and financial barriers that may shape the conditions within which institutions implement STEAM education.

The framework should therefore be interpreted as an integrative conceptual representation of the dimensions identified in the study rather than as a tested causal or explanatory model. It provides a structure for considering STEAM implementation within a broader organizational context and for examining how leadership practices, professional capacities, material conditions, and perceived barriers coexist within that context.

To make the integration of the quantitative and qualitative components explicit, Table 7 presents a joint display of the principal findings from both components. Because the two components addressed complementary rather than identical constructs, the joint display does not imply direct correspondence between quantitative indicators and qualitative categories. Instead, it shows how the qualitative findings contextualize and extend the quantitative evidence on resource provision within the broader organizational context of STEAM implementation.

Table 7. Joint display of quantitative and qualitative findings on organizational conditions associated with STEAM implementation.

As shown in Table 7, the integration of the two components revealed both convergence and complementarity. Material provision was relevant in both datasets: the quantitative findings showed variation in the provision levels of some specialized resources, while material resources and related barriers were prominent in leaders’ accounts. At the same time, the qualitative findings extended the quantitative evidence by identifying professional, organizational, and relational dimensions that were not measured in the national survey. The integrated findings therefore situate resource provision within a broader organizational context without implying causal relationships among the identified dimensions.

5 Discussion and conclusions

This study examined STEAM implementation in Lithuanian early childhood education institutions by integrating national-level quantitative evidence on the provision of teaching and learning resources with qualitative accounts of educational leaders concerning the broader organizational context of implementation. Taken together, the findings indicate that material resources constitute an important dimension of STEAM implementation, but that they coexist with human resources, professional learning, collaboration, leadership practices, learning environments, partnerships, and organizational and financial conditions. The findings therefore support a multidimensional understanding of STEAM implementation rather than an interpretation based on any single resource or organizational factor.

The quantitative findings revealed substantial variation in the provision levels of specific teaching and learning resources. Within the thematic area “Exploring and Understanding the Environment,” several resources showed provision levels above 70%, whereas specialized resources showed lower provision levels. Most notably, the national-level provision rate was 30.01% for STEAM laboratory kits designed for exploring natural and physical phenomena and 44.37% for prism sets. These findings indicate that the reported provision of some specialized resources was substantially below the identified level of need. However, resource provision alone cannot be interpreted as a direct measure of the extent or quality of STEAM implementation. STEAM education involves pedagogical decisions concerning how resources are incorporated into inquiry, experimentation, problem-solving, and interdisciplinary learning. This distinction is consistent with previous research emphasizing that educational technologies and materials acquire pedagogical value through their integration into meaningful learning practices rather than through their availability alone (Bybee, 2013; Deák and Kumar, 2024; Walshe et al., 2025).

The qualitative findings provide a broader organizational context for interpreting these quantitative patterns. Material resources and digital technologies constituted the most frequently identified resource subcategory in leaders’ responses (n = 27), while teachers’ STEAM competencies and understanding of the STEAM approach were also frequently identified (n = 23). Leaders additionally referred to learning environments, open-ended and natural materials, teachers’ creativity and initiative, external partnerships, and financial and project-based resources. Rather than supporting a conclusion that material resources are secondary to human resources, these findings suggest that leaders perceived STEAM implementation through multiple co-occurring resource dimensions. This interpretation is compatible with research that emphasizes the interaction between pedagogical competence, resources, learning environments, and institutional support in integrated STEAM education (Deák and Kumar, 2024; Margot and Kettler, 2019; Thibaut et al., 2018).

The leaders’ accounts also illustrate the relevance of professional learning and collegial collaboration within this broader organizational context. Sharing good practice and mentoring was the most frequently identified leadership-related subcategory (n = 21), followed by systematic professional development (n = 20) and efforts to strengthen STEAM within the organizational culture (n = 18). Leaders described peer learning, joint planning, mentoring, reflection on practice, seminars, training programmes, and institution-based professional learning. These findings resonate with the literature on professional learning communities, in which collaboration, reflective practice, and professional dialogue are regarded as important mechanisms of collective professional learning (Fullan, 2020b; Stoll et al., 2006). They are also consistent with STEAM research emphasizing the importance of sustained professional development that connects conceptual understanding with classroom practice (Deák and Kumar, 2024; Margot and Kettler, 2019; Movahedazarhouligh et al., 2023).

Leadership practices represented another dimension of the organizational context described by participants. Leaders referred to strategic planning, the integration of STEAM into institutional and curriculum documents, support for teachers’ professional growth, teacher motivation, and efforts to strengthen STEAM within organizational culture. These findings can be interpreted in relation to learning-centred and capacity-building perspectives on educational leadership, which emphasize leaders’ role in creating organizational conditions for professional learning, collaboration, and educational development (Fullan, 2020a, 2020b; Hallinger, 2011; Leithwood et al., 2020; Thornton, 2023). However, the present findings do not establish leadership as a causal or hierarchically dominant determinant of STEAM implementation. Rather, leadership practices were one of several organizational dimensions represented in leaders’ accounts.

The analysis of perceived barriers further reinforces the need for a multidimensional interpretation. Limited provision of specialized STEAM resources and digital technologies was the most frequently identified barrier subcategory (n = 17), followed by lack of dedicated STEAM learning spaces and laboratory equipment (n = 15). Pedagogical barriers related to teachers’ STEAM competencies and understanding were identified in the responses of 12 participants, while insufficient time for planning and collaboration (n = 9) and insufficient funding (n = 6) were also represented. A small group of participants (n = 4) reported no significant barriers or indicated that material limitations were not perceived as the primary barrier in their institutional context. These findings indicate that material constraints were prominent in leaders’ accounts, while pedagogical, organizational, and financial barriers were also represented. The findings therefore support an interpretation in which several types of barriers may coexist rather than one in which material constraints are treated as either the sole or a secondary barrier.

This multidimensional pattern is consistent with broader perspectives on educational change and organizational learning. Professional competence, opportunities for collaborative learning, time, institutional support, material provision, and leadership may be understood as components of an organizational context rather than as isolated determinants of educational practice (Fullan, 2020a; Mitchell and Sackney, 2011; Stoll et al., 2006; Thornton, 2023). In the context of STEAM education, this perspective is particularly relevant because inquiry-based and interdisciplinary practices require not only access to appropriate materials and technologies but also professional knowledge, opportunities for experimentation and reflection, and organizational arrangements that enable teachers to work collaboratively (Deák and Kumar, 2024; Thibaut et al., 2018; Walshe et al., 2025).

An additional insight concerns the learning environment and the nature of the resources identified by educational leaders. Alongside specialized technologies and equipment, participants referred to open-ended, natural, and everyday materials and to indoor and outdoor learning environments. This finding suggests that the material dimension of STEAM implementation should not be reduced to specialized or technologically advanced equipment. Previous research similarly associates early STEAM learning with flexible environments and opportunities for inquiry, exploration, experimentation, and child-initiated activity (DeJarnette, 2018; Walshe et al., 2025). At the same time, the comparatively low provision levels of some specialized resources in the quantitative data indicate that specialized equipment remains a relevant consideration and should not be displaced by an exclusive emphasis on pedagogical creativity.

External partnerships constituted a further, although less frequently identified, dimension of leaders’ accounts. Participants referred to collaboration with universities, museums, families, and other organizations as sources of resources and learning opportunities. Such partnerships may extend the resources available within individual institutions and contribute to professional and organizational learning. This finding is consistent with research emphasizing collaboration and external networks as components of institutional capacity for educational innovation (OECD, 2025a, 2025b; Thibaut et al., 2018; Walshe et al., 2025).

The integration of the quantitative and qualitative findings therefore provides a more differentiated interpretation than either component could provide independently. The quantitative component identifies national-level patterns and variation in reported resource provision levels, while the qualitative component situates resource-related issues within leaders’ accounts of professional competence, leadership practices, collaboration, learning environments, partnerships, and organizational constraints. The two components should not be interpreted as measuring identical constructs or as establishing causal relationships between resource provision and STEAM implementation. Instead, their complementarity situates material provision within a broader organizational context encompassing multiple dimensions.

On this basis, the data-informed conceptual framework presented in Figure 1 synthesizes the organizational dimensions identified across the study. It brings together leadership practices, professional learning and collaboration, human resources, material resources and learning environments, organizational and financial conditions, external partnerships, and perceived barriers in relation to STEAM implementation. The framework does not establish a hierarchy among these dimensions and should not be interpreted as a tested causal or explanatory model. Its theoretical contribution lies instead in organizing empirically identified dimensions within a coherent organizational perspective that may inform further research on STEAM implementation in early childhood education.

From a practical perspective, the findings suggest that efforts to support STEAM implementation should consider both material provision and broader organizational conditions. Investment in specialized STEAM resources, digital technologies, and appropriate learning environments remains relevant, particularly given the lower provision levels of some specialized resources identified in the national survey and the prominence of material barriers in leaders’ responses. At the same time, the qualitative findings indicate that educational leaders also associate STEAM implementation with systematic professional learning, collegial collaboration, strategic planning, and external partnerships. Rather than prioritizing one dimension at the expense of another, institutional and policy-level strategies may therefore benefit from considering these dimensions in combination.

6 Limitations and directions for future research

This study has several limitations. Although the mixed-methods design enabled the integration of national quantitative data with qualitative insights from educational leaders, the qualitative findings reflect only the perspectives of educational leaders. The qualitative component involved a voluntary-response sample of 38 educational leaders and was not intended to be statistically representative. This voluntary participation may have introduced self-selection bias, as leaders who chose to participate may have differed systematically from those who did not respond, for example in their level of interest in, engagement with, or experiences of STEAM implementation. Consequently, the qualitative findings may not capture the full range of perspectives among educational leaders and should be interpreted as reflecting the experiences and perceptions of the participating leaders rather than those of the broader population of early childhood education leaders. Furthermore, the results provide insight into leaders’ perspectives on organisational conditions but do not represent teachers’ classroom practices or children's experiences of STEAM education. In addition, information on institution size and participants’ years of leadership experience was not collected.

A further limitation concerns the trustworthiness procedures used in the qualitative analysis. Although both authors collaboratively reviewed and iteratively refined the coding structure and resolved interpretive differences through discussion, the full dataset was not coded independently by each author. Consequently, formal inter-coder reliability statistics, such as Cohen's kappa, were not calculated. Member-checking was also not conducted. These methodological choices should be considered when interpreting the qualitative findings, as the study relied primarily on collaborative researcher review rather than independent coding or participant validation.

The quantitative component was descriptive and focused on reported levels of teaching and learning resource provision. It did not measure the quality or extent of STEAM implementation and cannot establish causal relationships between resource provision and educational practice. Although 1,116 institutions were requested to provide data, responses were obtained from 831 institutions (74.46%); therefore, the quantitative findings describe patterns among participating institutions and should not be interpreted as representing complete national coverage.

The study was conducted within the Lithuanian early childhood education system; therefore, the findings should be interpreted with caution when applied to other educational contexts. The qualitative participants were not selected according to institutional resource-provision levels, and the quantitative and qualitative components addressed complementary rather than identical constructs. Accordingly, their integration provides a broader contextual interpretation but does not constitute a matched explanatory comparison between resource provision levels and organisational conditions.

A further limitation concerns the approximately six- to seven-month time lag between the quantitative and qualitative phases of the study. During this period, institutional conditions, resource availability, or STEAM-related practices may have changed. Therefore, the quantitative and qualitative data should not be interpreted as representing exactly the same point in time. This temporal gap may limit the extent to which the qualitative findings can be directly related to the resource-provision patterns identified in the earlier quantitative phase.

The study focuses on organisational conditions rather than educational outcomes. It does not examine relationships between organisational leadership, teachers’ instructional practices, and children's learning outcomes; therefore, the findings should not be interpreted as evidence of direct educational effects or causal mechanisms.

Future research should include multiple stakeholder groups, such as teachers, parents, and policymakers, and employ comparative or longitudinal designs. Integrating organisational data with classroom observations and children's learning outcomes could provide a more comprehensive understanding of the relationships between organisational capacity, STEAM implementation, and educational outcomes. Longitudinal research could also examine how these relationships develop over time.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Elizabeth Stosich, Fordham University, United States

Reviewed by: Desfa Yusmaliana, Bangka Belitung Muhammadiyah University, Indonesia

Hosam R. I. Badawy, Al Khawarizmi International College, United Arab Emirates

Baibhaw Kumar, University of Miskolc, Hungary