Christian Maximilian Thurn


Last Name

Thurn

First Name

Christian Maximilian

ORCID

0000-0002-5942-3273

Organisational unit

02871 - Didaktische Ausbildung

Search Results

Publications 1 - 10 of 44
  • How to develop and implement teaching projects in outdoor education
    Item type: Journal Article
    Thurn, Christian Maximilian; Zwyssig, Adrian; Gubelmann, Hanspeter; et al. (2025)
    ETH Learning and Teaching Journal
    Learning through projects can raise interest and motivation, and support the construction of competencies, disciplinary, and interdisciplinary knowledge via working on real-life problems in realistic settings. One form of project-based learning is outdoor education, that is, situating learning and instruction in settings outside the regular classroom. We present a course for students in the teacher education program at ETH Zurich that implements project-based education on two layers: the course itself is project-based, and the pre-service teachers create project-based outdoor teaching units during the course. We describe how we balance freedom and guidance, and scaffolding in the course design. In addition, we report how students respond to and evaluate our course, and discuss challenges and opportunities for lecturers. By presenting sample projects and insights from the implementation and continuous development of the project-based course, we aim to inspire and guide lecturers at ETH Zurich and other universities who consider implementing project-based courses in their teaching.
  • Improving the utility of non-significant results for educational research: A review and recommendations
    Item type: Journal Article
    Edelsbrunner, Peter Adriaan; Thurn, Christian Maximilian (2024)
    Educational research review
    When used appropriately, non-significant p-values have the potential to further our understanding of what does not work in education, and why. When misinterpreted, they can trigger misguided conclusions, for example about the absence of an effect of an educational intervention, or about a difference in the efficacy of different interventions. We examined the frequency of non-significant p-values in recent volumes of peer-reviewed educational research journals. We also examined how frequently researchers misinterpret non-significance to imply the absence of an effect, or a difference to another significant effect. Within a random sample of 50 peer-reviewed articles, we found that of 528 statistically tested hypotheses, 253 (48%) were non-significant. Of these, 142 (56%) were erroneously interpreted to indicate the absence of an effect, and 59 (23%) to indicate a difference to another significant effect. For 97 (38%) of non-significant results, such misinterpretations were linked to potentially misguided implications for educational theory, practice, or policy. We outline valid ways for dealing with non-significant p-values to improve their utility for education, discussing potential reasons for these misinterpretations and implications for research.
  • How to understand a research question-a challenging first step in setting up a statistical model
    Item type: Other Journal Item
    Edelsbrunner, Peter; Sebben, Simone; Frisch, Lisa K.; et al. (2023)
    Religion, Brain & Behavior
  • Optimizing the Learning Process in Higher Education: The Six Process Features of Learning
    Item type: Journal Article
    Thurn, Christian Maximilian; Daguati, Simona (2025)
    Zeitschrift für Hochschulentwicklung
    Human learning is characterised by six process features: learning is active, constructive, emotional, self-regulated, situated, and social. However, teaching often fails to adequately honour these features, particularly within higher education. To address this issue, we explain the six process features of learning and present examples of how to integrate them in class. As this article is intended for lecturers teaching at higher education institutions, we offer methodological guidance by providing additional material including visualizations, a didactic self-assessment tool, and posters that can be applied in diverse classroom settings, aiming to enhance teaching practices across disciplines.
  • Who makes use of prior knowledge in a curriculum on proportional reasoning?
    Item type: Conference Paper
    Nussbaumer, Daniela; Thurn, Christian Maximilian; Schumacher, Ralph; et al. (2017)
    Proceedings of the 39th Annual Meeting of the Cognitive Science Society
  • Sensitizing future teachers to psychological research on gender and STEM
    Item type: Journal Article
    Berkowitz Biran, Michal; Braas, Thomas; Thurn, Christian Maximilian (2022)
    ETH Learning and Teaching Journal ~ Diversity and Inclusion in Teaching and Learning
    What leads less women to pursue STEM careers? What does research find about differences in girls’ and boys’ educational trajectories? Students and faculty may have heard about gender bias, the leaky pipeline, gender stereotypes, or gender differences in the brain, but it is often difficult to grasp the underlying complexity of these topics. As social scientists in a technical university, we think that learning more closely about research in this field is helpful in developing a balanced and critical perspective. We have thus developed a course on gender issues in education and STEM for students in the teacher education program at ETH Zurich. In this paper, we first introduce some of the main issues in the context of gender and STEM, around which our course is designed. We then describe the pillars of our course. The course is interactive, with students presenting and critically discussing psychological and educational research. We walk students through the various controversies in the field: the nature-nurture question, gender differences vs. similarities, biases vs. interests, gender stereotypes and potential interventions. In a final assignment, students in small groups integrate several papers into a blog-post. Finally, we describe how students respond to our course, and discuss the challenges we as lecturers experience throughout.
  • Identifying central misconceptions via network analysis
    Item type: Other Conference Item
    Thurn, Christian Maximilian (2023)
    A core aspect of successful instruction is to identify and change students’ initial misconceptions about a topic. Assessing misconceptions and prior knowledge in physics is often done via concept inventories. These concept inventories are traditionally analyzed by summing up correct answers to infer on students’ conceptual knowledge. Yet, for the relational character of conceptual knowledge (e.g., Goldwater & Schalk, 2016), network analysis is more suitable to model the relations of students’ answers. When analyzing concept inventories through the lens of network analysis, researchers and educators can gain important insights into students’ knowledge structures and identify problematic misconceptions that hinder conceptual change. For this endeavor, I used data about concepts in magnetism. In total, 2210 students from primary and secondary school classes in Switzerland participated in instruction on magnetism in 15 cognitively activating lessons. They completed a concept inventory on magnetism before and after instruction. Following the approach by Brewe et al. (2016), I used network analysis on the distractor answers (the misconceptions) to analyze which misconceptions were central and maybe hindered learning, and which misconceptions persisted after instruction. The results showed that at pretest central misconceptions that related to many other misconceptions were that copper is magnetizable, or related to the shielding of a magnetic field via iron and wood plates. At posttest, central misconceptions were related to the shielding of magnetic fields via aluminum plates and about fair experiments. I will discuss how teachers could use such results to tackle central misconceptions.
  • A Model for Teaching for Conceptual Change? Six Challenges Regarding the ICAP Framework
    Item type: Other Conference Item
    Thurn, Christian Maximilian; Edelsbrunner, Peter; Berkowitz, Michal; et al. (2024)
    Conceptual Change in the Era of Digital Transformation: 13th International Conference on Conceptual Change - Programm & Abstracts
    The ICAP (Interactive, Constructive, Active, Passive) framework is based on constructive ideas of student learning and widely used by practitioners and researchers. The ICAP frame work links students’ overt behaviors to covert cognitive processes and learning outcomes and proposes that “higher” modes of engagement (I > C > A > P) are likely to increase learning. We theoretically discuss these assumptions of the ICAP framework in light of its po tential applicability to learning situations aimed at supporting conceptual change. We arrive at six challenges that question the framework’s applicability to conceptual change-focused instruction and further educational environments: 1) the equation of overt student behavior with covert cognitive processes, 2) the favorism of constructive and interactive modes, 3) the potential pitfall of having activities in the classroom that do not equate cognitive activation, 4) the weak and overstated empirical evidence in favor of the ICAP hierarchy, 5) the unclear guidance about when to attend to students’ products, and 6) the complex directive for prac titioners who would like to implement the ICAP framework. We aim to stimulate discourse on the multifaceted nature of learning processes and restate the importance of formative as sessment to unveil students’ learning at the covert cognitive level.
  • Change in Conceptual Understanding: The Role of Learning Opportunities, Prior Knowledge, and Intelligence
    Item type: Doctoral Thesis
    Thurn, Christian Maximilian (2021)
    The aim of this doctoral dissertation is to grasp the effect of prior knowledge to future learning as well as the relation between cognitive abilities and prior knowledge in physics within the framework of conceptual change. For this purpose, I conducted two studies, which are described in four empirical articles. The first study dealt with the impact of domain-specific prior knowledge in mathematics and physics as well as cognitive abilities onto the transfer of knowledge in proportionality, which was introduced via two isomorphic concepts. Theoretically, we hypothesized that the prior learning about the topic of floating and sinking, which implicitly embraces the proportional concept of density, has a positive influence on learning proportional reasoning. We assessed prior knowledge on the topics of density and speed, mathematical prior knowledge, and cognitive abilities. Contrary to our expectations, the results, however, showed that the mere participation in a teaching unit on floating and sinking did not have an effect on the transfer on proportional reasoning. In fact, prior knowledge in mathematics and cognitive ability played a bigger role. If, however, a test was used that assessed prior knowledge in physics as a continuous variable in contrast to the assessment of whether the teaching took place, there was a small effect of physics prior knowledge onto the mathematical concept of proportionality. The second study also concerned the relation of domain-specific prior knowledge and intelligence. This time, however, the topic was the physical concept of fields. I examined whether the prior learning about the topic of magnetism has a positive influence on learning the related topic of magnetic, electric and gravitational fields. To assess student’s knowledge and possible misconceptions I made use of several methods that capture conceptual understanding. Three empirical chapters of this thesis are based on this second study. In the first chapter, I investigated two central research questions of this doctoral dissertation: 1) How does prior knowledge affect conceptual learning in secondary school students in physics? 2) What is the relation between intelligence and prior knowledge during learning? I assessed conceptual understanding with three methods: a concept inventory, a speeded-reasoning paradigm, and concept mapping. I explored the effect of prior knowledge and intelligence and I could replicate results on the application of the speeded-reasoning paradigm. Prior knowledge had stable and sustainable effects on conceptual understanding assessed by the concept inventory and the speeded-reasoning paradigm in the pretest as well as in the posttest. At the same time, the learning gain was larger in the group without prior knowledge in magnetism. Thus, this group also profited from the instruction and the gap between the learners did not increase. Intelligence had positive effects especially in the group with existing prior knowledge, as intelligence affected the acquisition of prior knowledge, and had direct effects on posttest achievement, as detected by a mediation model. In the following chapter, which has already been published, I investigated how concept maps used across time can provide insights on conceptual change. In addition, I compared the knowledge networks of learners and experts on the topic of fields. I could discover that the knowledge structures not only grow across time but that they also change in terms of their structure: in the beginning, the structures fragmented more, to then form an integrated knowledge network at the end of instruction. Furthermore, I could identify concepts that are central to understand the fields concept. In the final empirical chapter, I could replicate central aspects of the previous chapter. Moreover, I examined differences between learners with and without prior knowledge in magnetism. In addition, this chapter addressed the systematic comparison of network indices that are commonly used to investigate concept maps. I found differences in knowledge structures across time, but barely between learners with and without prior knowledge in magnetism. Additionally, I could identify network indices that reliably show changes in knowledge structures. Overall, the studies in this doctoral dissertation helped to gain interesting results on the additional contribution of intelligence to conceptual understanding. Intelligence showed positive effects in particular when there was high prior knowledge. With regard to the strength of the effects, prior knowledge, however, showed larger effects than intelligence. This finding highlights again the importance of prior knowledge. Furthermore, I could replicate and extend basic results on conceptual change. The assumption of the co-existence of misconceptions and scientific conceptions was supported. Methodologically, the thesis added a contribution to the valid investigation of knowledge structures by means of concept maps and network analytic methods.
  • Students’ opinions on a digital tool for constructing and analyzing concept maps
    Item type: Other Conference Item
    Daguati, Simona; Thurn, Christian Maximilian (2024)
    Abstracts Congress of the German and Austrian Psychological Society, Vienna 2024
    Concept maps organize knowledge elements visually: They visualize concepts and their relations via nodes and links, while all relationships need to be justified by the learner. Concept maps stimulate constructive learning processes and capture knowledge gaps. They can therefore be utilised to support learning as well as to assess knowledge structures. A big challenge, however, lies in the analysis of concept maps. They are difficult to interpret and time-consuming to evaluate. Network analysis provides an objective and scalable approach to analyse concept maps. It addresses questions such as: Which concepts are well-connected and central? How does the aggregated concept map across learners look like? How do learners’ maps correspond to an expert map? To benefit from network analysis with concept maps in teaching, we developed the web-based R Shiny dashboard ConceptMappR, which will be freely available. We used the dashboard in three different courses at a Swiss university. To further refine the tool, we asked students to fill in a survey on its usability. Our research question was: How do students perceive the tool and which suggestions for improvement do they have? Answers from 25 participants showed that students viewed the tool as intuitive and powerful. A System Usability Score of 72.32 indicated an acceptable usability. A mean Net Promoter Score of 6.1 on a scale from 0 to 10 suggested that students would likely recommend the tool to colleagues or friends, although there is room for improvement.
Publications 1 - 10 of 44