Evaluating the Usability of Pervasive Conversational User Interfaces for Virtual Mentoring

摘要

To improve the academic and professional achievement of underrepresented minorities in computing, a newfound interest in innovative mentoring practices has captivated STEM education researchers. Studies suggest that virtual mentoring conversational agents can be leveraged across multiple platforms to provide supplemental mentorship, offsetting the lack of access to in-person mentorship in disadvantaged communities. A within-subjects mixed-method experiment was carried out to assess the usability of a mentoring conversational agent. Mobile interfaces (Twitter and SMS) were compared to each other and against a web-based embodied conversational agent (ECA). Results suggest that mobile interfaces are more usable than the web-based ECA. The findings from this study help to identify areas for improvement in virtual learning alternatives and other potential applications for pervasive conversational interfaces.