How connections matter: factors affecting student performance in STEM disciplines

The weakening academic performance of U.S. high school students and the declining influence of the U.S. economy have been studied extensively. Several studies have pointed to the declining interest of U.S. high school students in Science, Technology, Engineering, or Math (STEM) disciplines. This paper investigates the connectivity aspects that affect a high school student´s ability to excel academically. Using data from the National Center for Education Statistics (NCES), this paper analyzes the connections students have with their families, schools, friends, and community. Using these connections, this paper introduces an optimal prediction model named the Kalman-based Student Connectivity Model. By asking a student specific questions, this model can predict his/her cognitive reading and math performance on standardized exams. Further, this model can be used as a predictor on how modifying lifestyle choices or environmental factors may affect a student´s performance in the STEM disciplines.