printlogo
http://www.ethz.ch/index_EN.html
Institut für Verkehrsplanung und Transportsysteme, ETH Zürich
 
print
  
Deutsch English

e-citations

 

, Sasidharan, Lekshmi, Wu, Ke-Feng, Menendez, Monica, 2015 >>
Author(s): Sasidharan, Lekshmi, Wu, Ke-Feng, Menendez, Monica
Title: Exploring the application of latent class cluster analysis for investigating pedestrian crash injury severities in Switzerland
Abstract: One of the major challenges in traffic safety analyses is the heterogeneous nature of safety data, due to the sundry factors involved in it. This heterogeneity often leads to difficulties in interpreting results and conclusions due to unrevealed relationships. Understanding the underlying relationship between injury severities and influential factors is critical for the selection of appropriate safety countermeasures. A method commonly employed to address systematic heterogeneity is to focus on any subgroup of data based on the research purpose. However, this need not ensure homogeneity in the data. In this paper, latent class cluster analysis is applied to identify homogenous subgroups for a specific crash type-pedestrian crashes. The manuscript employs data from police reported pedestrian (2009-2012) crashes in Switzerland. The analyses demonstrate that dividing pedestrian severity data into seven clusters helps in reducing the systematic heterogeneity of the data and to understand the hidden relationships between crash severity levels and socio-demographic, environmental, vehicle, temporal, traffic factors, and main reason for the crash. The pedestrian crash injury severity models were developed for the whole data and individual clusters, and were compared using receiver operating characteristics curve, for which results favored clustering. Overall, the study suggests that latent class clustered regression approach is suitable for reducing heterogeneity and revealing important hidden relationships in traffic safety analyses.2
Published in: Accident Analysis and Prevention
Volume: 85
Pages: 219 - 228
ISSN: 0001-4575, 1879-2057
Publication date / Date received: 2015-12-01
Publication status: Amsterdam
Publication status: Published
Subjects: Binary logit, Cluster analysis, Latent class, Pedestrian, Receiver operating characteristic (ROC) curve, Severity, Switzerland
Language: English
Keyword: Binary logit, Cluster analysis, Latent class, Pedestrian, Receiver operating characteristic (ROC) curve, Severity, Switzerland
DBID source: FORM-1445672996, SCOPUS-84944062678, WOS-000364884800021
DOI: 10.1016/j.aap.2015.09.020
Nebis system number: 000041481
, Sasidharan, Lekshmi, Wu, Ke-Feng, Menendez, Monica, 2015 >>
Author(s): Sasidharan, Lekshmi, Wu, Ke-Feng, Menendez, Monica
Title: Exploring the application of latent class cluster analysis for investigating pedestrian crash injury severities in Switzerland
Abstract: One of the major challenges in traffic safety analyses is the heterogeneous nature of safety data, due to the sundry factors involved in it. This heterogeneity often leads to difficulties in interpreting results and conclusions due to unrevealed relationships. Understanding the underlying relationship between injury severities and influential factors is critical for the selection of appropriate safety countermeasures. A method commonly employed to address systematic heterogeneity is to focus on any subgroup of data based on the research purpose. However, this need not ensure homogeneity in the data. In this paper, latent class cluster analysis is applied to identify homogenous subgroups for a specific crash type- pedestrian crash. The manuscript employs data form police reported pedestrian (2009-2012) crashes in Switzerland. The analyses demonstrate that dividing pedestrian severity data into seven clusters helps in reducing the systematic heterogeneity of the data and to understand the hidden relationships between crash severity levels and socio-demographic, environmental, vehicle, temporal, traffic factors, and main reason for the crash. The pedestrian crash injury severity models were developed for the whole data and individual clusters, and were compared using receiver operating characteristics curve, for which results favored clustering. Overall, the study suggests that latent class clustered regression approach is suitable for reducing heterogeneity and revealing important hidden relationships in traffic safety analyses.2
Publication date / Date received: 2015-01-01
Publication status: Published
Event name: 94th Annual Meeting of the Transportation Research Board (TRB 2015)
Event date: 11-15 January 2015
Place: Washington, D.C.
Language: English
DBID source: FORM-1454149592
, Sasidharan, Lekshmi, Menendez, Monica, 2014 >>
Author(s): Sasidharan, Lekshmi, Menendez, Monica
Title: Partial Proportional Odds Model
Subtitle: An Alternate Choice for Analyzing Pedestrian Crash Injury Severities
Abstract: The conventional methods for crash injury severity analyses include either treating the severity data as ordered (e.g. ordered logit/probit models) or non-ordered (e.g. multinomial models). The ordered models require the data to meet proportional odds assumption, according to which the predictors can only have the same effect on different levels of the dependent variable, which is often not the case with crash injury severities. On the other hand, non-ordered analyses completely ignore the inherent hierarchical nature of crash injury severities. Therefore, treating the crash severity data as either ordered or non-ordered results in violating some of the key principles. To address these concerns, this paper explores the application of a partial proportional odds (PPO) model to bridge the gap between ordered and non-ordered severity modeling frameworks. The PPO model allows the covariates that meet the proportional odds assumption to affect different crash severity levels with the same magnitude; whereas the covariates that do not meet the proportional odds assumption can have different effects on different severity levels. This study is based on a five-year (2008–2012) national pedestrian safety dataset for Switzerland. A comparison between the application of PPO models, ordered logit models, and multinomial logit models for pedestrian injury severity evaluation is also included here. The study shows that PPO models outperform the other models considered based on different evaluation criteria. Hence, it is a viable method for analyzing pedestrian crash injury severities.2
Published in: Accident analysis and prevention
Volume: 72
Pages: 330 - 340
ISSN: 0001-4575
Publication date / Date received: 2014-11-01
Publication status: Amsterdam
Publication status: Published
Subjects: Crash severity, Partial proportional odds model, Ordered logit model, Multinomial logit model, Pedestrian, Comparison
Language: English
Keyword: Crash severity, Partial proportional odds model, Ordered logit model, Multinomial logit model, Pedestrian, Comparison
DBID source: FORM-1407825418, WOS-000343843700034
DOI: 10.1016/j.aap.2014.07.025
Nebis system number: 000041481
, Sasidharan, Lekshmi, Menendez, Monica, 2014 >>
Author(s): Sasidharan, Lekshmi, Menendez, Monica
Title: Partial Proportional Odds Model - A Better Choice for Analyzing Pedestrian Crash Injury Severities
Abstract: The conventional methods for crash injury severity analyses include either treating the severity data as ordered (e.g. ordered logit/probit models) or non-ordered (e.g. multinomial models). The ordered models require the data to meet proportional odds assumption, according to which the predictors can have only the same effect on different levels of the dependent variable, which is often not the case with crash injury severities. On the other hand, non-ordered analyses completely ignore the inherent hierarchical nature of crash injury severities. Therefore, treating the crash severity data as either ordered or non-ordered results in violating some of the key principles. To address these concerns, this paper explores the application of a partial proportional odds (PPO) model to bridge the gap between ordered and non-ordered severity modeling frameworks. The PPO model allows the covariates that meet the proportional odds assumption to affect different crash severity levels with the same magnitude; whereas the covariates that do not meet the proportional odds assumption can have different effects on different severity levels independently. This study is based on the five year (2008-2012) national pedestrian safety dataset for Switzerland. A comparison between the application of PPO models, ordered logit models, and multinomial logit models for pedestrian injury severity evaluation is also included in the study. The analysis shows that PPO models outperform ordered logit and multinomial logit models based on different evaluation criteria used in the paper.
Publication date / Date received: 2014-01-01
Publication status: Zurich
Publication status: Published
Language: English
DBID source: FORM-1395927281
, Sasidharan, Lekshmi, Menendez, Monica, 2014 >>
Author(s): Sasidharan, Lekshmi, Menendez, Monica
Title: Beneficial and Detrimental Factors Influencing Pedestrian Crash Injury Severities in Switzerland Using Partial Proportional Odds Model
Abstract: Problem: According to the Swiss Microcensus, Switzerland is a place where people choose to walk more than 40% of their daily trip time, resulting in a high pedestrian crash to total crash ratio when compared to many other developed countries. Furthermore, the high number of old and young pedestrians as well as the large number of pedestrian crashes with higher levels of severity makes the pedestrian crash analysis for Switzerland very important. However, to the best of author's knowledge, no scientific research has been conducted on evaluating the role of critical factors influencing the severity level of pedestrian crashes in the country. This study aims to identify the important crash contributing factors that alleviate and aggravate the injury severity levels of pedestrian-vehicle crashes in Switzerland. Method: This study employs a Partial Proportional Odds model (Generalized ordered logit model) to analyze pedestrian crash injury severities in Switzerland, based on the five-year (2008-2012) pedestrian safety data. The study also focuses on a separate analysis of influential factors for crashes involving old and young pedestrians, who are more vulnerable to higher levels of injury severity when involved in a crash . Results: The results of the study indicate that old pedestrians and young pedestrians are more prone to serious injuries when involved in crashes. Irrespective of age, dark unlighted areas, sight obstructions, mid-block crossings and heavy vehicles were found to adversely affect the injury severity levels of pedestrians. On the other hand, German speaking areas, low speed urban areas, and flat road sections were found to positively influence pedestrian injury severities in Switzerland. Practical Applications: Discussion on the effects of pedestrian, driver and infrastructure characteristics, and roadway crossing behavior of pedestrians in different age groups will help to improve traffic safety related policies in Switzerland.
Published in: SVT Working Paper
Volume: 71
Publication date / Date received: 2014-01-01
Publication status: Zürich
Publication status: Published
Subjects: Pedestrian, Injury, Severity, Ordered logit, Partial proportional odds, Generalized ordered logit, Proportional odds assumption, Parallel lines assumption, Switzerland
Language: English
Keyword: Pedestrian, Injury, Severity, Ordered logit, Partial proportional odds, Generalized ordered logit, Proportional odds assumption, Parallel lines assumption, Switzerland
DBID source: FORM-1406983855
, Sasidharan, Lekshmi, Donnell, Eric T., 2014 >>
Author(s): Sasidharan, Lekshmi, Donnell, Eric T.
Title: Propensity scores-potential outcomes framework to incorporate severity probabilities in the Highway Safety Manual crash prediction algorithm
Abstract: Accurate estimation of the expected number of crashes at different severity levels for entities with and without countermeasures plays a vital role in selecting countermeasures in the framework of the safety management process. The current practice is to use the American Association of State Highway and Transportation Officials' Highway Safety Manual crash prediction algorithms, which combine safety performance functions and crash modification factors, to estimate the effects of safety countermeasures on different highway and street facility types. Many of these crash prediction algorithms are based solely on crash frequency, or assume that severity outcomes are unchanged when planning for, or implementing, safety countermeasures. Failing to account for the uncertainty associated with crash severity outcomes, and assuming crash severity distributions remain unchanged in safety performance evaluations, limits the utility of the Highway Safety Manual crash prediction algorithms in assessing the effect of safety countermeasures on crash severity. This study demonstrates the application of a propensity scores-potential outcomes framework to estimate the probability distribution for the occurrence of different crash severity levels by accounting for the uncertainties associated with them. The probability of fatal and severe injury crash occurrence at lighted and unlighted intersections is estimated in this paper using data from Minnesota. The results show that the expected probability of occurrence of fatal and severe injury crashes at a lighted intersection was 1 in 35 crashes and the estimated risk ratio indicates that the respective probabilities at an unlighted intersection was 1.14 times higher compared to lighted intersections. The results from the potential outcomes-propensity scores framework are compared to results obtained from traditional binary logit models, without application of propensity scores matching. Traditional binary logit analysis suggests that the probability of occurrence of severe injury crashes is higher at lighted intersections compared to unlighted intersections, which contradicts the findings obtained from the propensity scores-potential outcomes framework. This finding underscores the importance of having comparable treated and untreated entities in traffic safety countermeasure evaluations.
Published in: Accident analysis and prevention
Volume: 71
Pages: 183 - 193
ISSN: 0001-4575
Publication date / Date received: 2014-01-01
Publication status: Amsterdam
Publication status: Published
Subjects: Severity probability, Highway safety manual, Propensity score, Potential outcome, Causal model, Lighting
Language: English
Keyword: Severity probability, Highway safety manual, Propensity score, Potential outcome, Causal model, Lighting
DBID source: FORM-1406984055, WOS-000340304000021
DOI: 10.1016/j.aap.2014.05.017
Nebis system number: 000041481

 

 

 

Wichtiger Hinweis:
Diese Website wird in älteren Versionen von Netscape ohne graphische Elemente dargestellt. Die Funktionalität der Website ist aber trotzdem gewährleistet. Wenn Sie diese Website regelmässig benutzen, empfehlen wir Ihnen, auf Ihrem Computer einen aktuellen Browser zu installieren. Weitere Informationen finden Sie auf
folgender Seite.

Important Note:
The content in this site is accessible to any browser or Internet device, however, some graphics will display correctly only in the newer versions of Netscape. To get the most out of our site we suggest you upgrade to a newer browser.
More information

© 2016 ETH Zurich | Imprint | Disclaimer | 29 January 2016
top