We are currently witnessing the growth of the maker movement. According to the Hackerspaces organization and the U.S. Census Bureau, there has been a 14-times increase in the number of maker spaces worldwide over the last 10 years. One of the most compelling aspects of open-ended learning environments such as Makerspaces and Fabrication Labs is that they integrate formal STEM concepts with social emotional, 21st century skills. In these environments, students learn complex technical skills in teams, create unique artifacts, and develop self-directed, creative solutions to problems in their communities. Such learning experiences develop essential skills that are not well taught in traditional school curricula that have a significant impact on students’ ability to adapt to the challenges of the future. In particular, we are seeing a sharp decrease in the type of “routine” jobs that require workers to follow a predefined procedure. Jobs that require workers to complete “non-routine” tasks, think outside the box, work with others, and solve complex problems are becoming increasingly difficult to fill. Makerspaces and digital fabrication labs can prepare the next generation of students to learn how to invent new procedures for solving unique problems, rather than simply applying already mastered procedures.
However, capturing evidence of the development of 21st century skills remains a challenge in makerspaces. Papavlasopoulou et al. (2017) reported that 85% of the empirical studies were either purely qualitative or utilized mixed methods. Timotheou (2019) writes “we now have enough evidence of the value of computation making, allowing for scaling-up the impact and measurement via quantitative studies” (2019, p. 227).
In short, we need a better way to observe the mechanisms and processes of social emotional learning inside a makerspace. Student learning in open-ended learning environments like the makerspace often happen at varying directions and paces, largely while students engage in self-directed, unsupervised work. This project is about using multimodal learning analytics, such pose detection and gaze inferrence algorithms, to capture learning processes in open-ended learning environments, and use that data to support learning and teaching (Figure 1).
Figure 1: Our makerspace is equipped with 8 high resolution cameras to extract multimodal data (e.g., 3D poses, gaze direction) from videos. On the right, feed from one of the cameras shows two students working (and sharing visual attention) together, while another student is at the laser cutter station at the far wall. On the left, the same scene is represented by data captured through our system, showing the position and gaze of the three students mapped onto a floorplan of the makerspace.
References
-
Papavlasopoulou, S., Giannakos, M. N., & Jaccheri, L. (2017). Empirical studies on the Maker Movement, a promising approach to learning: A literature review. Entertainment Computing, 18, 57–78.
-
Timotheou, S., & Ioannou, A. (2019). On Making, Tinkering, Coding and Play for Learning: A Review of Current Research. In D. Lamas, F. Loizides, L. Nacke, H. Petrie, M. Winckler, & P. Zaphiris (Eds.), Human-Computer Interaction INTERACT 2019 (pp. 217–232). Cham: Springer International Publishing.