Posters
The poster session takes place on Wednesday, September 10 before lunch (10:30) with the following accepted submissions:
From Curricula to Classrooms: Primary School Teachers' First Experiences and Challenges in Teaching CS
Felix Ziemann and Torsten Brinda
[Poster] [Text]Traffic lights, robot vacuum cleaners or the first smartphone – in today's society even young children grow up in an increasingly digital networked world and experience digital systems in their everyday lives. To empower them as active participants and contributors rather than mere passive consumers in this digital world, education must equip them with the necessary competencies from an early age. In some federal states in Germany, selected computer science (CS) competencies have already been integrated into the primary school curricula. These include topics such as first programming skills as well as a basic understanding of computer systems. However, primary school teachers have not yet been systematically prepared to teach these contents. This poses a significant challenge, particularly for older teachers, who presumably had little or no contact with CS during their own school education either. This paper presents the development and design of an online survey conducted among primary school teachers. The survey explores various aspects, such as teachers' beliefs about CS, their current teaching practices regarding the consideration of CS competencies, as well as their CS-related attitudes, growth mindset, and self-efficacy. In addition, the teachers are also asked about their individual preferences for professional development offers. The insights gained from the study will be used to develop interventions, including teaching materials and professional development courses. Initial results from the pilot study are still pending, but will be shown on the poster.
Novice Programming Misconceptions in Elementary Education
Monika Mladenović and Žana Žanko
[Poster] [Text]Programming misconceptions have been studied since the 1980s, when the first significant scientific papers on the topic emerged. Although much has changed since then—particularly in the technology surrounding us—novice programming misconceptions have remained largely consistent, regardless of age, programming language, or time period. However, most existing studies have been conducted at the university level, and there is still a lack of research focused on younger learners. As programming becomes an increasingly integral part of Informatics and Computer Science curricula in elementary schools, it is important to investigate misconceptions among young learners. Over the past decade, we have conducted four studies with fifth- and sixth-grade students, focusing on misconceptions related to the text-based programming language Python and the procedural programming paradigm. These studies were carried out in real classroom settings, involving a total of 435 students across four school years, 25 classes, five schools, and five teachers. In this paper, we present 17 identified programming misconceptions related to basic programming constructs: variables, sequencing, conditionals, and loops.
Conceptual Models for Teaching Quantum Computing: A Normative Analysis
Daniel Krosse and Alexander Best
[Poster]As quantum computing gains relevance in computer science education, there is a growing need for instructional models that address its conceptual complexity. This study presents a normative analysis of the quantum bit and quantum entanglement, deriving conceptual models from disciplinary and educational literature. The models aim to support learners in developing coherent and scientifically grounded mental representations, while also addressing common misconceptions. The findings offer a foundation for future empirical research and the design of effective teaching strategies in quantum computing education.
K-6 Teachers' Wishes and Expectations for Computer Science Training
Eve Tessenow and Alexander Best
[Poster]We explore K-6 teachers' expectations and wishes for computer science teacher training. n = 8 primary and secondary school teachers in Germany were interviewed using participatory research methods (sketching, mind mapping). Data were analyzed using qualitative content analysis, checked per intra- (.61) and intercoder agreement (.41 to 1.00). We outline the research methodology and presents initial results.
Engaging Learners through a Physical Computing-Based Chess Escape Room (Best Poster)
Deividas Roščenkovas, Anita Juskeviciene, and Gabrielė Stupurienė
[Poster]Scientific research has proven that playing chess can help to develop analytical and creative thinking skills. However, chess is not an attractive way to spend time for most people of all ages. This work presents a case of conveying chess through an experience close to escape rooms. Escape room elements are widely used in various fields in order to enhance people's involvement in a certain activity. A chess escape room system was designed, describing its functionality in detail and justifying the choice of components used. In the future, an improved version of this system could be applied in the non-formal education process, and due to its extensive feedback functionality, it is also suitable for use with children with special needs.
PyToPseu: Automatic Natural-Language Formulations of Programming Constructs to Avoid Misconceptions
Jean-Philippe Pellet and Patrick Wang
[Poster] [Text]Introduction to programming remains one of the first delicate topics in computer science education. In this poster proposal, we describe our efforts to equip beginners with a tool which, from Python code, generates a line-by-line natural-language description of the code. This tool is designed to help students understand programming constructs and avoid common misconceptions. The tool is not based on LLMs, but aims to stay very close to the actual code and is based on a set of rules that map Python constructs to their natural-language interpretations. We believe that this approach can enhance the learning experience for beginners, especially those who may struggle with understanding the syntax and semantics of basic programming constructs.
Assessing Design Thinking in Computing: Towards an Automated Evaluation of Skills
Snieguolė Bagočienė
[Poster]Design thinking (DT) is increasingly recognized as a core competence in 21st-century computing education, encouraging creativity, critical thinking, and collaboration. Yet, assessing such skills remains time-consuming, subjective, and difficult to scale. Advances in artificial intelligence (AI), natural language processing (NLP), and computer vision (CV) present new opportunities for automating and enhancing complex educational assessments. This study introduces a theoretical informatics engineering model for an AI-based system designed to assess DT competences in K-12 computing education. The proposed system integrates validated creativity constructs (fluency, flexibility, originality, elaboration), semantic divergence measures, and critical thinking indicators. It processes both textual and visual artefacts using NLP engines (e.g., SBERT, GPT-based models), CV tools (e.g., YOLOv8), and transformer-based scoring algorithms. The architecture is modular, interoperable with learning management systems (LMS), and capable of generating real-time, personalized feedback through adaptive dashboards. Though currently at the design stage, this is the first AI-based framework for DT assessment in the Lithuanian K-12 context. By combining computational creativity theory, semantic models, and scalable automation, the system lays the groundwork for evidence-based, high-fidelity assessment practices in education. This model represents a novel informatics engineering solution to the persistent challenge of evaluating higher-order thinking skills in digital learning environments.
“Art Heist at the Museum”: Data Science Education through Educational Breakouts in an Informatics Teaching and Learning Lab
Kira Klaner and Nils Pancratz
[Poster] [Text]In an era of increasing digitalisation, data literacy is essential for responsible participation in society. However, key data science skills such as data collection, exploration and data-driven modeling and evaluation remain underrepresented in current curricula and teacher training. Educational Escape Rooms (EERs) offer a promising way to address this gap by embedding technical content into immersive, narrative scenarios. This paper presents an EER on Data Science, developed within the teacher education programme at the Informatics Teaching and Learning Lab at the University of Hildesheim. Framed as a fictional crime investigation, the EER uses the open-source data mining tool Orange to introduce supervised and unsupervised machine learning techniques, including decision tree classification and clustering. The scenario also fosters critical reflection on algorithmic bias through the use of AI-generated images and deliberately stereotyped character profiles.