Investigating Underage Alcohol-Intoxicated Driver Crash Patterns in Louisiana

Abstract

Despite the zero tolerance and minimum legal drinking age laws since 1999, crashes caused by underage drinking drivers have occurred every year in the United States, which is a huge cause for concern for roadway safety. This study utilized multiple correspondence analysis (MCA) on 9 years (2010 to 2018) of at-fault, underage (aged 15 to 20 years), alcohol-intoxicated driver crashes in Louisiana to investigate hidden crash attribute patterns. This exploratory multivariate analysis technique identifies systematic associations among categories of qualitative variables rather than recognizing the effect of a single factor on the response variable. The results exhibited the capability of MCA in discovering the meaningful clouds of crash contributory factors from a complex multidimensional dataset. Fatal crashes happened to underage alcohol-intoxicated drivers who were not using seatbelts on high-speed curve segments under dark without streetlights, whereas single-vehicle crashes resulted in moderate injuries among drinking novice teenagers operating light trucks. The findings also revealed the driving behavior patterns of rookie drivers after alcohol intake that resulted in collisions. For example, male teenagers were engaged in impaired driving during weekends under adverse weather conditions, and underage drinking drivers tended to use cellphones during late-night driving. Targeting critical attributes identified from associations could be helpful in reducing the number of related crashes and fatalities. Furthermore, knowledge gained about the attribute groups identified in this study could be included in educational training programs targeting risky driving maneuvers. Integration of multiple interventions could be more strategic in minimizing underage drinking collisions.

A Clustering Regression Approach to Explore the Heterogeneous Effects of Risk Factors Associated with Teen Driver Crash Severity

“…Female teenagers are less likely to be involved in injury collisions ( 11 ), but the related crashes are more likely to be rear-end ( 12 , 13 ). Because of the minimum legal drinking age (MLDA) and zero-tolerance laws, teenagers are less exposed to intoxicated driving than adults ( 14 ). Teen drivers with a blood alcohol concentration (BAC) of 0 mg/dL have a similar crash likelihood in respect of young adults with a BAC of 0.05 to 0.079 mg/dL ( 15 ).…”

Investigation on Hazardous Material Truck Involved Fatal Crashes Using Cluster Correspondence Analysis

“…A low-dimensional space displays the associations between the column elements and row elements in the data matrix, and the positions of the row and column points are consistent with their associations in the table. Many researchers have applied different forms of correspondence analysis in transportation safety studies, such as taxicab correspondence analysis [24,30], multiple correspondence analysis [31][32][33][34][35][36], and joint correspondence analysis [37,38]. As one of the unsupervised machine learning approaches, correspondence analysis does not require any distribution hypothesis among data.…”

“…A previous study has identified its ability to efficiently reduce dimensionality and compile results into easy-to-read plots for in-depth crash analysis [32]. However, distinguishing different clusters depends on subjective judgment, which relies on the categorical variable feature approximation results in a low-dimensional space [36]. In recent years, some transportation safety studies used cluster correspondence analysis due to its ability to discover the underlying cluster structures and reduce multicollinearity among data [24,[39][40][41][42].…”

Using Cluster Correspondence Analysis to Explore Rainy Weather Crashes in Louisiana

“…In recent years, multiple correspondence analysis (MCA) has become a popular unsupervised algorithm tool to analyze categorical variables in exploring crash databases. For example, previous research studies investigated crash patterns through association knowledge of fatal run-off crashes ( 28 ), pedestrian crashes ( 29 , 30 ), wrong-way crashes ( 31 , 32 ), underage alcohol crashes ( 33 ), and so forth. It has been argued that the MCA is a better unsupervised machine-learning approach to tackle difficulties in other unsupervised approaches such as optimizing measurable thresholds in association rule mining ( 28 ).…”

“…It has been argued that the MCA is a better unsupervised machine-learning approach to tackle difficulties in other unsupervised approaches such as optimizing measurable thresholds in association rule mining ( 28 ). Although the MCA performs efficient dimensionality reductions and compiles results into presentable plots, one of the limitations of this approach is that researchers have to rely on a subjective approach to identify clusters based on the proximity of categorical variable attributes on the low-dimensional approximation ( 33 ). By combining the cluster analysis and correspondence analysis, the CCA provides specific clusters by partitioning individual attributes based on the profiles over the categorical variables through low-dimensional projection and reduction of multicollinearity of the crash dataset in the process ( 34 ).…”