Emergence in the U.S. Science, Technology, Engineering, and Mathematics (STEM) workforce: an agent-based model of worker attrition and group size in high-density STEM organizations
Authored by Ronald Iammartino, John Bischoff, Christopher Willy, Paul Shapiro
Date Published: 2016
DOI: 10.1007/s40747-016-0015-7
Sponsors:
No sponsors listed
Platforms:
NetLogo
Model Documentation:
ODD
Model Code URLs:
Model code not found
Abstract
The United States government forecasts a shortage of 1,000,000 Science, Technology, Engineering, and Mathematics (STEM) workers over the next 10
years, putting STEM workforce sustainability at risk. The U.S. federal
government has launched a range of programs, initiatives, and
commissions to address this critical shortage. Past research has shown
that organizational group-oriented culture, smaller work team groups, and worker homogeneity relate to lower attrition rates and higher levels
of reported worker satisfaction. Applying Complex Adaptive Systems, this
study takes a systems approach to examine worker attrition rates and
worker group sizes in relation to STEM density across high-density STEM
organizations. The base-case organization for this study is the National
Aeronautics and Space Administration (NASA) because it maintains the
highest density of STEM workers across all U.S. federal organizations
and consistently ranks first in teamwork, innovation, and worker
satisfaction in the annual Office of Personnel Management (OPM) Federal
Employee Viewpoint Survey. The methodology for this study comprises an
empirically validated NETLOGO agent-based model of worker attrition
using OPM data sets for high-density STEM organizations. Model
initialization parameters are represented and validated by historical
attrition data for NASA and two control group organizations. The control
groups include a lower-limit high-density STEM model and a medium
high-density STEM model for validation with historical worker attrition
data for the Environmental Protection Agency and Federal Communications
Commission. The findings for this study confirm that STEM worker density
is negatively related to attrition rate across all high-density
organization types. The model output observations further show the
emergence of a negative relationship between organizational STEM density
and average worker group size, yet the opposite association is observed
for STEM workers. The agent-based modeling approach is an important
addition to the current line of academic focused research on STEM
workers because it provides a bottom-up insight which helps inform
theories and policy effects on ways to mitigate forecasted STEM
shortages. Future research could be extended to apply Tipping Point
Theory to STEM density and attrition rate variability to better
understand STEM threshold ranges across high-density STEM organizations.
Future models could also investigate specific STEM and Non-STEM worker
characteristics to include age, gender, salary, or education level.
Tags
systems