SBIR Phase II: An embedded and in-context professional learning platform for math problem-solving instruction

This project proposes to develop an innovative approach to improve and sustain math educators' problem solving teaching skills. Despite the expectations placed on math teachers by the Common Core State Standards, most are insufficiently prepared to teach students how to become critical thinkers. Much of this problem is due to limited pedagogical skills of teachers in providing adequate problem solving instruction and supports on top of teachers' own limited problem solving skills. This project remedies this with its integrated modules and powerful analytics engine that suggests learning pathways for both expert and novice teachers. They anchor their research in National Council of Teacher's of Mathematics Principles to Action. It will help teachers develop confidence and skills in planning and evaluating their lessons, as well as understanding student misconceptions and intervening in a timely manner. Teachers who approach problem solving with confidence inspire students to approach difficult math tasks the same way. This has great implications for how many students will continue to enroll in Science, Technology, Engineering and Math programs. In addition, the project sets the stage for educators to develop 21st Century skills including critical thinking, communication and collaboration - essential job skills for the young minds they mentor. This effort refines and scales up their product, which is a web and mobile application that works seamlessly in conjunction with our current student-facing platform, to provide teachers with timely supports for improving students' problem-solving skills and math communication. This project will deliver professional development continuously and in-context using virtual peers, rich rubrics, interactive tools and actionable data. The analytics engine leverages adaptive learning models in order to build robust modules. The Data Collector Layer will contain interfaces for users to get recommendations, receive user feedback and provide other analysis reports. The Analytics Core Layer will be implemented using a collection of machine learning algorithms. The Service Layer will calculate recommendations based on user profile, user feedback, pre-stored best practices and other use cases. The Persistence Layer will store and get calculated data to recommendation engine's own database. The company plans to conduct several formative evaluations during the course of the project, as well as two pilot studies at the end of each year with a control and experiment group. The results will enable them to determine the effectiveness of ongoing, just-in-time supports for improving teachers' skills and confidence inside and outside the classroom.