Text Mining Analysis for Assessing Washington Accord Graduate Attribute Profiles through Techno-Socio Project-Based Learning Program

Techno-socio Project-Based Learning programs (PBLs) are a very effective instruction method to nurture engineering students’ skillsets and mindsets as future professionals by identifying, analyzing, and solving issues through group work with a wide variety of teammates. This is a highly practical learning experience for participants to communicate with industry professionals, local government officers, and other local people about societal issues and possible technology applications. This program especially puts strong emphasis on experiential learning “outside the classroom,” namely field research activities that can supplement the limitations of conventional lecture style studies. Engineering students can absorb what they need through practical experiences together with teammates from different backgrounds. While these experiences are valuable, in most cases they are not quantitatively measurable.Meanwhile, PBLs are highly suitable for achieving the 11 Graduate Attribute Profile (GAP) skills and awareness essential for global engineers defined by the Washington Accord (WA). They can be acquired through real-world experiences, as “practice makes perfect.” However, these GAPs contain many unmeasurable factors. Moreover, even in cases in which GAPs are acquired, it is difficult to clarify where and how the acquisition happened. This article examines an experiment to identify the causality between the techno-socio PBL contents and learning outcomes related to this Washington Accord 11 Graduate Attribute Profile (WA11GAP) by applying a text-mining technique. The results conclude that this methodology is useful not only for grasping the effectiveness of PBL program contents from a cause-effect perspective but is also applicable to other nonstandard teaching methods that cannot be quantitatively assessed with conventional exams.