Fostering Algorithmic Thinking within a Productive-Failure-Based Workshop Utilizing AI

This research examines the promotion of algorithmic thinking - a future skill for K-12 students to understand the digital world - through AI education and block-based programming tools like Scratch. With the increasing emphasis on computational thinking, it is now included in many school curricula. AI tools, such as the Teachable Machine, open new avenues for improving computational and algorithmic problem-solving skills, although the most effective classroom methods for developing these skills remain under-researched. Through a mixed-methods case study, this research explores the development of algorithmic thinking, identifies failures in students' problem-solving processes, and examines the structure of problem-solving within a workshop designed around the productive failure approach. This workshop, conducted by pre-service teachers in a Computer Science teaching-learning lab, uses AI in Scratch to promote algorithmic thinking. Following a productive failure approach, students attempt to solve problems before receiving targeted instruction (PS-I). The goal is to provide students with a better understanding and learning environment that enhances their algorithmic thinking skills. The sample consists of N = 23 secondary school students. Their algorithmic skills were assessed before and after their participation. N = 7 groups (14 students) of them are recorded and the solving process within the steps of algorithmic thinking and the productive failures made are analyzed by means of qualitative content analysis. The results suggest that a short workshop is not sufficient to develop algorithmic thinking, with most failures occurring in the algorithm development phase. This study provides insights into the design of instructional settings that better support algorithmic thinking skills.