SBIR Phase II: Increasing Student Engagement through Adaptive Instructional Video Delivery

This SBIR Phase II project addresses low student engagement through a dynamic video delivery environment with a novel, real-time algorithm that adjusts the instructional pathway to each student's readiness level. According to research, the greatest challenge facing teachers is overcoming the fact that between 25-66% of students are disengaged. Adapting instruction to each student's learning background has long been touted as the most effective method to drive student engagement; when instruction is individualized, engagement and outcomes increase dramatically. By calibrating video segments to student readiness, this project delivers appropriately challenging instruction to most effectively sustain engagement. Building on promising results in Phase I, the project aims to improve engagement and outcomes for all Science, Technology, Engineering and Math (STEM) students and, most pointedly, impacts English Language Learners (ELL) who often fall behind in STEM classes because instruction is beyond their readiness and in-class support is insufficient. Considering that ELL population is the fastest-growing population of public school students in the United States, this project has a significant potential to drive a STEM proficient workforce. Furthermore, given a strong, pre-existing client base, the project team is well positioned to penetrate the $110 million immediately addressable market. By 2020, the project team expects to be changing how 10% of U.S. middle schoolers absorb STEM instruction, opening a pathway to the adoption of powerful blended classroom techniques that enhance engagement, improve learning, and promote STEM careers. The core innovation is a research-founded algorithm to deliver appropriately complex instruction through all video segments. For successful delivery of appropriately complex instructional video segments, the project (1) measures student complexity readiness, (2) maintains a database of video segments categorized by complexity (using factors such as syntactic and lexical complexity), and (3) uniquely unifies that information to identify a real-time next step for that student (which video segment should be shown). The aim is to enhance outcomes by providing each student an appropriately complex instructional pathway forward. With roots in the leading interactive video solution, this project is made possible through access to a large database of tagged and curated videos (hundreds of thousands) and questions (millions) and provides an unprecedented opportunity to develop adaptive instructional pathways for a diverse learning spectrum. For Phase II, the project team's goal is to deploy a fully functioning classroom prototype of a product built around a middle school Physical Science unit and measure impact on engagement and learning. The team will employ stimulated recall interviewing techniques, speak aloud interview protocols, and quantitative usage logs to assess the feasibility of highly-adaptive video instruction to increase student engagement. Evaluation will include both formative components (to gauge implementation and iterative improvement) as well as summative components to assess impact on engagement and learning outcomes in a mixed-methods design including multiple pilot studies with control and experiment groups. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.