SBIR-STTR Award

This SBIR Phase I project will provide an interactive learning environment for educational sketching which is critical in many Science, Technology, Engineering, and Math (STEM) fields. Sketching is an integral part of education from elementary school arithmetic to high school physics, and is used in professions ranging from medicine to engineering. Sketching is part of spatial visualization training, which has been shown to increase graduation rates, especially among women and other underrepresented minorities. Until now, sketching assignments have been done with pencil and paper. This project will bring sketching assignments to touchscreens on tablets and computers. The algorithm developed will provide students with real-time feedback based upon their preliminary attempts, and teachers will be able to see exactly where students are having difficulty. A unique advantage of this approach is that it tracks student persistence and rewards students who stick with a sketching assignment until completed correctly. This is an advantage over most online education that relies on multiple-choice questions, which can only be attempted once. The project will initially focus on spatial visualization training for college engineering students, but ultimately will extend throughout STEM and be adapted for K-12. The increase in retention of students in STEM will aid in meeting the national targets to support economic growth.The software developed will automatically grade student sketches and provide feedback similar to an expert teacher. When a student makes an error, personalized hints are generated. By providing incremental help and rewarding students when they try on their own, student persistence is encouraged. The sketching will be performed on touchscreens, and the software will be compatible with iOS, Android, Windows, Chrome, and Linux operating systems. There are significant challenges with grading sketches, especially when compared to multiple-choice questions. If the grading algorithm is too lenient or too strict, students will lose confidence in the software. The grading algorithm will consider the sketched image, the difficulty of the assignment, and will utilize sketching data from early trials for fine-tuning purposes. Personalized assignments that gradually increase in difficulty as needed, as well as interactive 3D graphics and animations will be provided. A teacher interface will be developed that shows each student's progress down to the detail of each sketch attempt. Additionally, a metric will be developed to quantify student persistence based upon reliance on hints in order to identify which students need more individual mentoring. The scope of this project is the development of the grading algorithm, user interface, assignment and tutorial content, and teacher interface.

This SBIR Phase II project will develop a personalized learning environment for guiding students in effective use of educational sketching in Science Technology Engineering and Math (STEM). There is an estimated need for 1 million more STEM professionals in the U.S. over the next decade. However, the dropout rate for students in STEM majors ranges from 48% to 69%. Recent reform in mathematics education in the U.S. has focused on sketching and modeling so that students have a better understanding of the problems they are solving rather than rote memorization. Sketching is also important for improving the ability to visualize three dimensional shapes in engineering, medicine, and science. Teaching sketching has been shown to increase retention in STEM by over 50% among students entering college with low spatial visualization skills, especially among women and other underrepresented minorities. However, the traditional method of sketching on paper requires teaching expertise and classroom time, which has limited its use. Our touchscreen software will improve upon sketching on paper, by providing immediate feedback and intelligent hints when students are stuck. A key advantage of our approach is that it can reward persistence by having students retry a sketching assignment until they get it right, as opposed to multiple-choice questions that can only be tried once. Impact will include improved conceptual understanding in K-6 mathematics and STEM education at all levels.This project will develop an intelligent learning environment where students can freehand sketch on a touchscreen phone, tablet, or computer. They will receive personalized feedback based on the sketches they draw to improve their conceptual understanding and promote persistence. The software algorithms developed will automatically grade student sketches and provide hints that are just enough to keep the student engaged but not so much as to give away the solution. An expert teacher knows that developing skill and increasing persistence requires that students struggle a bit as they overcome obstacles to learn new concepts. The algorithm for interpreting student sketches and providing hints will mimic an expert teacher looking over the shoulder of the student. By interpreting student intent in their sketches, the algorithm will be able to distinguish between intentional marks and unintentional ones that are due to poor coordination or sloppiness. Subject areas include elementary school mathematics, spatial visualization for 7-12th grade technical education, post-secondary vocational and engineering programs, and eventually physics, geology, medicine and other STEM areas. This project will replace sketching on paper and pencil with widely available touchscreens, and make educational sketching more effective and engaging.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.