System dynamics provides a valuable framework for analyzing the complexities of a system over time. Using causal loop diagrams (CLDs), this study identifies two feedback mechanisms in the context of integrating computational thinking (CT) into education, focusing on leverage points such as coding-centric curricula and teacher competence development. The modeling shows how reinforcing and balancing feedback loops shape system behavior, demonstrating that interventions like coding-based approaches can deliver immediate benefits but risk hindering the long-term development of broader CT skills applicable to interdisciplinary problem-solving. The study enhances the analysis by uncovering patterns emerging from system dynamics, including the constraints limiting growth, reliance on symptomatic solutions, unintended consequences of quick fixes, and the prioritization of successful areas at the expense of others. These patterns highlight the importance of addressing foundational issues, such as teacher training and comprehensive curriculum design, to avoid overshadowing other critical goals of CT applications in interdisciplinary fields. By synthesizing CLDs, the study showcases the circular interactions between variables and the diverse dynamics influencing CT education, offering insights into potential scenarios and storylines for integrating CT into education systems.

BACKGROUND: In today’s world, acquiring essential skills is crucial for empowering individuals, particularly children, to handle everyday challenges and tasks in a technologically advanced society. Among these skills, computational thinking (CT) plays a vital role in problem-solving and adapting to the complex and evolving demands of the 21st century. However, it is necessary to explore the role of other thinking skills alongside CT, considering that CT cannot be improved and applied in isolation.

OBJECTIVE: This paper aims to address the gap in knowledge regarding the application of systems thinking to CT development and its integration into education settings.

METHODS: Results from two studies, focusing on CT development using educational robotics and maker technologies, form the basis of this paper. The research findings are synthesized and consolidated using the systemic FMA (framework of ideas, methodology, and area of concern) model.

RESULTS: The research findings illustrate that utilizing a diverse set of approaches, methods, and tools can improve CT skill development across different educational settings.

CONCLUSIONS: The adapted FMA model promotes methodological pluralism and facilitates a critical examination of CT development boundaries, leading to both conceptual and practical changes. This approach enables the recognition of emergent properties, the design of interventions, and the incorporation of multiple perspectives.

Computational thinking (CT) is a fundamental cognitive skill that provides a problem-solving approach applicable across various domains. Integrating CT into K–12 education is crucial for developing the competencies that students need; however, its implementation presents significant challenges. While most research on CT has focused on specific areas and stakeholders, a more systematic and holistic approach is essential for successful integration. This paper, using a systems thinking theoretical lens, explores the complexities and interdependencies of this process, informed by 5 years of empirical research. Through the development of a detailed rich picture illustrating the integration, we identified four key dimensions as leverage points within the system: context and settings, stakeholders, theoretical and pedagogical strategies, and tools and approaches. These dimensions collectively shape the educational system for CT. This paper enriches the understanding of this integration by highlighting these leverage points, setting the stage for the application of additional systems thinking tools for integrating CT into education.

This study explores the conceptualization of an idealized design for computational thinking (CT) education from teachers’ perspectives, guided by the principles of interactive planning. While CT education has received significant attention, the possibility of envisioning its desired state remains underexplored. Through workshops and discussions with teachers, this study follows an idealized design approach. It involves identifying challenges in the current state of CT education, examining worst-case scenarios if these challenges persist, and exploring the possibility of envisioning a desired current state for CT education. The findings reveal not only a lack of understanding of CT but also other challenges such as inequitable access to resources and insufficient collaboration among educators. Additionally, although potential worst-case scenarios and ideal CT education are complex to conceptualize, key themes emerged. Worst-case scenarios included a widening achievement gap, diminished problem-solving skills, and reduced motivation among advanced learners. Meanwhile, the envisioned ideal CT education encompassed themes such as skills and competencies, equity and access, and the transformative concept of School 2.0. These themes were subsequently positioned within cognitive, situated, and critical framings of CT to better contextualize the findings and connect them with existing research. This framing helped illustrate the coherence of these themes while broadening the scope of CT education research.

Computational Thinking (CT) is increasingly recognized as a critical 21st-century skill, yet its integration into education has revealed a range of conceptual, pedagogical, and systemic challenges. These include the absence of a unified definition, the tendency to equate CT with programming, insufficient teacher training, and unclear implementation strategies. This thesis employs a systems thinking approach to critically examine the complexities involved in integrating CT into education, focusing on the Swedish K–12 education.

The study argues that traditional approaches to CT education are often reductionist, addressing isolated components without accounting for the broader systemic interactions. Systems thinking offers an alternative by analyzing interdependencies, feedback loops, and emergent behaviors within educational systems. CT education is conceptualized here as a complex, interconnected system involving multiple stakeholders, institutional structures, and sociocultural dynamics. The thesis utilizes systems thinking techniques, including rich pictures and causal loop diagrams, to map and analyze the systemic forces shaping CT education. These tools are used to model challenges such as unclear curricular expectations, inconsistent teacher competencies, policy misalignments, and societal pressures for digital transformation.

Three questions guide the research. What characterizes the current state and key challenges of CT education from a systemic perspective? What system dynamics influence the integration of CT into education? How can CT education be envisioned and planned to reflect its complexity and an idealized present? To address these questions, the study adopts a qualitative methodology grounded in interactive planning and idealized design. Empirical data were collected through workshops with teachers, which facilitated the identification of systemic obstructions, the projection of reference scenarios, and the co-creation of idealized visions for CT education. Together, these efforts produced a theoretically grounded and empirically informed vision of CT education that moves beyond current limitations. The findings suggest that effective and sustainable CT integration requires a shift from reactive or preactive approaches toward interactive planning, which actively involves stakeholders in shaping a desirable present.

This thesis makes two key contributions. Theoretically, it positions systems thinking as a central analytical lens for research on CT education. Methodologically, it introduces a structured framework, rooted in idealized design, that educators, policymakers, and researchers can use to navigate complexity and support long-term transformation in CT integration.

Computational Thinking (CT) pervasively shares its methods, practices, and dispositions across other disciplines as a new way of thinking about problem-solving. Few studies have been carried out studying CT from an Information Systems (IS) perspective. This study elaborates on how systems thinking (ST), an acknowledged theory in the IS field, bonds to CT to address some well-known common issues related to CT such as reductionism and dogmatism and to supplement the computing nature of CT with behavioral and societal facets involved in its implications. We studied how ST is applied to CT research in the literature. To do so, two primary approaches have been identified that link ST and CT. First, ST is embedded in CT practices meaning that ST is considered as a component of CT. Second, ST and CT are parallelly studied, and ST is considered as a supplementary concept to CT. Correspondingly, we propose a complementary approach that looks at CT from the ST lenses to provide a clearer picture of CT in an educational context. Moreover, we expect this new perspective can help to broaden the development of educational CT concepts and scenarios by including new notions such as framework, interpretation, norms, paradigm, and context.

In today's rapidly changing world, the acquisition of essential skills is crucial for the success of young individuals. Among these skills, computational thinking (CT) plays a vital role in problem-solving and adapting to the complex and evolving demands of the 21st century. However, there is a need to explore the integration of other thinking skills alongside CT, as well as their application in educational settings.

This study aims to address the gap in knowledge regarding the application of systems thinking to the development of CT and its integration into education. The primary objective is to explore the relationship between systems thinking and CT, providing a contextual framework for existing studies that focus on systems thinking in relation to CT. Additionally, the study explores how systems thinking can be applied to CT within educational contexts. By incorporating a systems thinking approach, a broader examination of the various factors involved in CT, including the technological landscape, individual skills and knowledge, and the social and cultural context, can be achieved.

The thesis comprises three papers that describe research efforts conducted over three years. These projects focused on CT development using educational robotics and maker technologies, aiming to build and enhance CT skills among individuals of different ages and perspectives. The findings of the research efforts are synthesized and consolidated using the systemic FMA model, a comprehensive model that interconnects the frameworks of ideas, methodology, and the area of interest. This model conceptualizes CT practices as a system encompassing emergent properties, multiple perspectives, design interventions, and social and ethical considerations. The adopted FMA model enables methodological pluralism and facilitates critical examination of the boundaries of CT development, leading to conceptual and practical changes.

The research contributes to the field of CT by providing insights into its theoretical foundations and practical applications, informing and guiding educational practices that are associated with CT. 

In many countries worldwide, Computational thinking (CT) is now considered as a fundamental skill for dealing with the challenges of the 21stcentury society. One of the most common ways of imparting CT knowledge in K-12 education is by teaching programming and coding, as it requires applying a set of concepts and practices that are essential for thinking computationally. However, learning to program can be challenging and it may take time to develop these skills in the context of school activities. Thus, complementing formal K-12 education with after-school or other types of informal learning activities aimed at fostering CT concepts and practices among young students can be an alternative approach to develop these skills. During the summer of 2021, we carried out a series of workshops in the context of a summer camp taking place at a public library, organized by a local municipality in southern Sweden. These workshops (with a total teaching duration of 20 hours in one week) consisted of activities where children aged 11-14had to assemble wheeled robots and then program them using a visual language to make them execute different types of tasks and challenges. The outcomes of our study show that roughly one third of the participants managed to program the robots with code that made use of CT core concepts, such as conditionals, loops, and logical operators, among others. The rest of the children did not manage to successfully apply these concepts and thus they could only manage to program sequential linear scripts. We argue that learning to program and understanding some of the main concepts, which are for the most part very abstract, is a process that takes time and thus, extracurricular activities can be an effective method to complement formal education and help young students develop their CT and programming skills.

Digital competence is a skill associated with the 21- century abilities essential to contribute to today’s and tomorrow’s digital and technical environments. Computational Thinking (CT), which is a thought process for problem-solving, is one of the emerging trends that make up digital competence. In our explorative study, we have used educational robotics with four pre-service teachers during their four- week s placement at different preschools. We applied three distinct and complementary approaches to design and conduct this study: Systems Thinking (ST); Technological Pedagogical Content Knowledge; and Computing Pedagogy. Our findings are categorized in two main perspectives: pre-service teachers and children. In the pre-service teachers' perspective, the participants indicated that their educational program lacks specific content and activities related to digital competence, CT, and programming. Despite the initial pre-service teachers’ thoughts on improvement of children's CT concepts, the findings show that CT practices such as collaboration and trial and error were developed. From the children’s perspective, the empirical findings illustrate that digital competence and CT development vary depending on the age of the children; whereas logical thinking and pattern recognition are skills that were present along the whole age range of children (ages 2-6), other CT skills like algorithmic thinking were developed among older children only (aged 5-6). We learned that an ST approach can behelpful, as multiple factors are involved in the practice. It reveals the underlying features of the situation that emerge when components of the system interact with each other.