Date of Award

Spring 5-2019

Degree Name

Master of Science (MS)

Advisor(s)

SUSAN TORTOLERO EMERY, PHD

Second Advisor

ROBERT EMERY, DRPH

Third Advisor

JOSE-MIGUEL YAMAL, PHD, MA

Abstract

College and university campuses contain a diverse set of potential health and safety risks mixed with a population that is equally varied. Such a combination requires effective risk management programs adapted to address these challenges. Methods currently exist to predict risk financing premiums and institutional risk control resources (inputs), but there are no models addressing relationships to final Environment Health & Safety (EHS) program outcomes (outcomes such as injuries or illnesses). Publicly available data obtained from the University of Texas System Office of Risk Management describing their 14 campuses was combined with publicly available data from the National Science Foundation to produce a model assessing the relationship between EHS resourcing inputs and EHS program outcomes. EHS program outcomes were represented by the outcome variable of workers’ compensation insurance modifiers and were compared to resourcing variables such as number of EHS full time employees, total net assignable square footage, research net assignable square footage, and research & development expenditure. When assessed individually, all resourcing variables revealed a negative linear relationship with the outcome variable (lower experience modifiers being indicative of better outcome performance). When using a multivariable stepwise estimation regression, all input variables were eliminated from the model except for research and development expenditure that presented the same negative correlation. The results of this exploratory study suggest that increased EHS resourcing is associated with improved EHS program outcomes. However, the limited sample size affected the statistical significance of the regression models and resulting interpretation. Future opportunities should be taken advantage of by other university systems to expand upon these preliminary findings and validate the observed correlations.

Share

COinS