A simple agent-based social impact theory model of student STEM selection

There is a growing body of knowledge describing the economic and social challenge faced by the United States because of the small (14%) and decreasing number of students pursuing Science, Technology, Engineering, and Mathematics (STEM) majors. We propose a simple two-period, agent-based simulation based on social impact theory to predict the % yield of STEM majors. The model indicates that changes with minimal (if any) cost could more than double the STEM yield. For example, allocating the STEM-oriented teaching talent in the first two years rather than in the last two years could increase yields by approximately 5.5%. Also, dividing or segregating students based on STEM orientation could increase yield by over 10%. We begin by briefly reviewing the literatures about STEM and social impact theory. Next, we describe our proposed model and numerical experiments using standard design of experiments methods. Finally, conclusions and suggestions for future research are provided.