Understanding Crash Patterns and Secondary Task Engagement Patterns of At-risk Drivers: A Comparison of Secondary Task Types and an Analysis of Naturalistic Driving

José Antonio Calvo IV

Major Professor: William S Helton, PhD, Department of Psychology

Committee Members: Yi-Ching Lee, Tyler Shaw, Carryl Baldwin

Online Location, Online
November 20, 2020, 01:00 PM to 02:00 PM


In 2016 there were over 34,000 fatal crashes in the US. Based on miles driven the two most vulnerable road users are younger and older drivers. Understanding how similar or dissimilar the crashes of these groups are can help inform how to decrease the number of these crashes.  Distraction is one factor that negatively effects crash likelihood.  This dissertation uses the Second Strategic Highway Research Program (SHRP2)’s Naturalistic Driving Study (NDS), which classifies over 40 specific secondary task that constitute distraction.  The first study seeks to identify the best way to condense these tasks into smaller categories. The first method used to condense the secondary tasks is based on multiple resource theory.  The second method used to condense groups of secondary tasks was a cluster analysis of driver reaction delay, using a K-means cluster of actual crashes.  It is hypothesized that when using a number of clusters equal to the number of groups in the multiple resource theory grouping the cluster method should look similar if not identical to the multiple resource theory grouping.  Furthermore, we’ll look at break down of secondary task engagement by age for both clustering types and compare differences in engagement patterns. 

The second study compares age, secondary task engagement (based on the previous study), maneuver judgement, and gender and their effect on crash likelihood using a hierarchical log-linear model.  The data used for this analysis will again be that of the SHRP2 NDS.