Building Trust in Robots in Robotics-Focused STEM Education under TPACK Framework in Middle Schools

Abstract

Abstract Recent research has examined effective use of robotics as a pedagogical tool in STEM education through the lens of TPACK (technological- pedagogical-content knowledge). Such a framework can help visualize and understand abstract content knowledge in a tangible and concrete manner, enrich kinesthetic learning, improve excitement and engagement of students, promote active learning, intrinsically and extrinsically motivate teachers and students, and thus enhance overall learning environment and outcomes. Not surprisingly, in recent years, application of robotics in STEM education has witnessed increased interest, become an area of active research, and attracted significant efforts for incorporating robotics into STEM curricula. Application of robotics in middle school STEM education is appropriate since the age and maturity level of students demand a greater emphasis on consideration of situated cognition, cognitive apprenticeship, collaborative learning, and inquiry-based learning frameworks, all of which can be addressed through the robotics. Nonetheless, based on our experience, we believe that despite its tremendous potential robotics cannot be widely incorporated into STEM education curricula until teachers and students develop trust in the robotics. Here, trust of teachers and students in robotics indicates their willingness to believe in the solutions provided by robots and to rely on the contributions of robots in STEM teaching and learning. Human trust in robots is extensively explored in fields such as service, manufacturing, etc. However, trust of teachers and students in robots in robotics-focused STEM education has not been studied yet. This paper is based on our collaboration with 20 teachers of 8 urban, inner-city schools and observations of over 200 students in their robotics-focused STEM lessons under TPACK framework. Using appropriate questionnaire techniques, we develop a 'trust vocabulary' that contains what the teachers and the students mean by trust in the robots for their lessons, and what factors and features of the robotics may affect their trust. Next, we develop a qualitative trust assessment method using a Likert scale and derive a quantitative trust computational model. We compare the qualitative and quantitative trust measurements and validate the quantitative trust model. We propose several hypotheses and investigate whether there are statistically significant differences in trust in robots between teachers and students, teaching subjects such as science and math, and participants' genders. We analyze whether the level of trust of the teachers in robotics effects their pedagogy and the level of trust of the students effects their learning methods. Based on the trust assessment results, we conduct a survey with the teachers and students and offer recommendations that may help enhance the trust levels of the teachers and students in the robotics for STEM education. The results are novel that may advocate using robotics as a trust-worthy pedagogical tool under the TPACK framework, argue incorporating robotics into STEM education curricula in middle schools, and help maintain appropriate levels of trust of teachers and students in robots, which may increase the overall learning outcomes.