The CTSkills App -- Measuring Problem Decomposition Skills of Students in Computational Thinking
Project Overview
The document discusses the integration of generative AI in education, highlighting the CTSkills App, which specifically targets the development of problem decomposition skills in computational thinking for K-12 students. This app addresses a critical gap in computational thinking education by emphasizing problem decomposition, a skill essential for effective problem-solving, particularly in the era of generative AI. The web-based assessment tool has been tested with students in grades 4-9, allowing educators to collect valuable data on students' proficiency in this area. The findings reveal insights into how understanding problem decomposition can enhance students' capabilities in navigating complex challenges and leveraging generative AI technologies. Overall, the document underscores the potential of generative AI to transform educational practices by providing innovative assessment tools that foster essential skills in the digital age.
Key Applications
CTSkills App
Context: K-12 education, specifically for students in grades 4-9
Implementation: Developed as a web application using JavaScript, HTML, and CSS; tested in classroom settings with 75 students across various grades.
Outcomes: Improved understanding of students' decomposition skills, ability to assess CT competencies, and automated data collection for educational research.
Challenges: Definition and assessment of decomposition skills remain unclear; variations in language comprehension among students; reliance on cross-sectional data limits longitudinal insights.
Implementation Barriers
Definition and Assessment Challenge
Lack of consensus on how to define and assess problem decomposition skills in computational thinking education.
Proposed Solutions: Development of a standardized assessment tool (CTSkills) to measure decomposition skills and establish baseline proficiencies.
Comprehension Variability
Variations in language comprehension among students could affect their understanding of task instructions.
Proposed Solutions: Incorporate language options and clearer instructions in future app revisions.
Data Limitations
Reliance on cross-sectional data limits the ability to establish causal relationships or track longitudinal developments.
Proposed Solutions: Future research should employ longitudinal designs to monitor individual student progress over time.
Project Team
Dorit Assaf
Researcher
Giorgia Adorni
Researcher
Elia Lutz
Researcher
Lucio Negrini
Researcher
Alberto Piatti
Researcher
Francesco Mondada
Researcher
Francesca Mangili
Researcher
Luca Maria Gambardella
Researcher
Contact Information
For information about the paper, please contact the authors.
Authors: Dorit Assaf, Giorgia Adorni, Elia Lutz, Lucio Negrini, Alberto Piatti, Francesco Mondada, Francesca Mangili, Luca Maria Gambardella
Source Publication: View Original PaperLink opens in a new window
Project Contact: Dr. Jianhua Yang
LLM Model Version: gpt-4o-mini-2024-07-18
Analysis Provider: Openai