Learning to See Like an Engineer: Capturing Shifts in Students' Perceptions Using Grounded Learning Analytics

• In Event: Measuring Making: Methods, Tools, and Strategies for Capturing Learning, Participation, and Engagement in Maker Activities

Abstract

Makerspaces are inherently messy learning environments. In them, students learn a variety of cognitive, metacognitive and social skills. This poster details a new method for capturing these changes though a combination of task-based assessments and qualitative, computational analysis. Using these methods, we found that students who took part in a maker workshop became more like engineers in their ability to reason about and solve complex problems. More specifically, the students learned to recognize the functional affordances of complex mechanisms—that wheels are for rotating and gears are for meshing. This shift in perspective is an important, empowering educational outcome, and provides new motivation for studying the educational impact of fostering a maker mindset in youth.

The current study was designed to answer three questions about students’ participation in a maker workshop. First, do students think more like engineers as a result of taking part in a long-term maker workshop? Second, is there a way to reliably and efficiently capture this change? And third, if this change in thinking does occur, can we look more closely at the data and begin to understand the specific ways in which the students’ thinking changes? To answer these questions, we designed a task-based assessment where students had to rebuild a gear system, fix everyday devices, and debug a microcontroller. These tasks were designed to assess skills that participants learned during the workshop, such as constructing a 3D object from 2D parts, building an input-output system, and understanding that complex systems are made of small interconnected parts.

In this poster, we focus on the gear assembly tasks. We developed an extensive coding scheme to estimate students’ performance and found that participants significantly improved from pre- to post-test. This suggests that the workshop had a positive effect on their ability to reason about and assemble complex mechanisms. Additionally, we asked experts (graduate students in mechanical engineering) to complete the same tasks and found that novices became more like experts in their perception of the salient features of the problem and in their problem-solving approach. This finding suggests that novices’ improvement actually reflects one (or several) skills that experts have gained through many hours of studying and interacting with mechanical systems. It makes us more confident that we are capturing an important aspect of what constitutes an engineer. This work builds on prior work that analyzed a single task (Davis, Schneider, & Blikstein, 2017). We extend this work by analyzing a second task and introducing analysis of the students’ gesture and body position using Kinect (motion capture) data – with the goal of designing proxy for the coding schemes we developed in this paper.

Authors

• Richard Davis, Stanford University

• Bertrand Schneider, Harvard University

• Paulo Blikstein, Stanford University