Analysis of naturalistic driving event data

Omitted-variable bias and multilevel modeling approaches

Paul P. Jovanis, Jonathan Aguero-Valverde, Kun Feng Wu, Venkataraman Shankar

Research output: Contribution to journalArticle

26 Citations (Scopus)

Abstract

Naturalistic driving studies have been conducted over the past 5 years or more and have commonly reviewed video and kinematic data to identify and analyze crash, near-crash, and critical-incident events. But statistical methods that are applicable to these event data are needed. This paper addresses two issues in model development for naturalistic driving event data: the test for omitted-variable bias and the exploration of the advantages of hierarchical model structures in data analysis. With roadway departure event data from the 100-Car Naturalistic Driving Study conducted at Virginia Tech Transportation Institute, Blacksburg, Virginia, logit models were used to estimate the probability that a crash or a near crash would occur, rather than a critical incident. The models indicated a substantial omitted-variable bias for estimation of the effect of context variables but little difference for driver variables. These tests indicated that modeling of naturalistic event data should have included variables that described the attributes of the event, the driver, and the context to reduce the likelihood of bias. Hierarchical model structures offer the advantage of driver-level predictors to parameterize the effects of event attributes and contexts. The models thus reflect how driver decisions are executed: drivers with particular characteristics (one level) find themselves in contexts in which they execute specific driving maneuvers (second level), which lead to certain outcomes. Suggestions for further research include testing with additional data sets and potential applications to analysis of crash surrogates.

Original languageEnglish (US)
Pages (from-to)49-57
Number of pages9
JournalTransportation Research Record
Issue number2236
DOIs
StatePublished - Dec 1 2011

Fingerprint

Model structures
Statistical methods
Kinematics
Railroad cars
Testing

All Science Journal Classification (ASJC) codes

  • Civil and Structural Engineering
  • Mechanical Engineering

Cite this

Jovanis, Paul P. ; Aguero-Valverde, Jonathan ; Wu, Kun Feng ; Shankar, Venkataraman. / Analysis of naturalistic driving event data : Omitted-variable bias and multilevel modeling approaches. In: Transportation Research Record. 2011 ; No. 2236. pp. 49-57.
@article{511ae4b327004120ad72a579e5eed46c,
title = "Analysis of naturalistic driving event data: Omitted-variable bias and multilevel modeling approaches",
abstract = "Naturalistic driving studies have been conducted over the past 5 years or more and have commonly reviewed video and kinematic data to identify and analyze crash, near-crash, and critical-incident events. But statistical methods that are applicable to these event data are needed. This paper addresses two issues in model development for naturalistic driving event data: the test for omitted-variable bias and the exploration of the advantages of hierarchical model structures in data analysis. With roadway departure event data from the 100-Car Naturalistic Driving Study conducted at Virginia Tech Transportation Institute, Blacksburg, Virginia, logit models were used to estimate the probability that a crash or a near crash would occur, rather than a critical incident. The models indicated a substantial omitted-variable bias for estimation of the effect of context variables but little difference for driver variables. These tests indicated that modeling of naturalistic event data should have included variables that described the attributes of the event, the driver, and the context to reduce the likelihood of bias. Hierarchical model structures offer the advantage of driver-level predictors to parameterize the effects of event attributes and contexts. The models thus reflect how driver decisions are executed: drivers with particular characteristics (one level) find themselves in contexts in which they execute specific driving maneuvers (second level), which lead to certain outcomes. Suggestions for further research include testing with additional data sets and potential applications to analysis of crash surrogates.",
author = "Jovanis, {Paul P.} and Jonathan Aguero-Valverde and Wu, {Kun Feng} and Venkataraman Shankar",
year = "2011",
month = "12",
day = "1",
doi = "10.3141/2236-06",
language = "English (US)",
pages = "49--57",
journal = "Transportation Research Record",
issn = "0361-1981",
publisher = "US National Research Council",
number = "2236",

}

Analysis of naturalistic driving event data : Omitted-variable bias and multilevel modeling approaches. / Jovanis, Paul P.; Aguero-Valverde, Jonathan; Wu, Kun Feng; Shankar, Venkataraman.

In: Transportation Research Record, No. 2236, 01.12.2011, p. 49-57.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Analysis of naturalistic driving event data

T2 - Omitted-variable bias and multilevel modeling approaches

AU - Jovanis, Paul P.

AU - Aguero-Valverde, Jonathan

AU - Wu, Kun Feng

AU - Shankar, Venkataraman

PY - 2011/12/1

Y1 - 2011/12/1

N2 - Naturalistic driving studies have been conducted over the past 5 years or more and have commonly reviewed video and kinematic data to identify and analyze crash, near-crash, and critical-incident events. But statistical methods that are applicable to these event data are needed. This paper addresses two issues in model development for naturalistic driving event data: the test for omitted-variable bias and the exploration of the advantages of hierarchical model structures in data analysis. With roadway departure event data from the 100-Car Naturalistic Driving Study conducted at Virginia Tech Transportation Institute, Blacksburg, Virginia, logit models were used to estimate the probability that a crash or a near crash would occur, rather than a critical incident. The models indicated a substantial omitted-variable bias for estimation of the effect of context variables but little difference for driver variables. These tests indicated that modeling of naturalistic event data should have included variables that described the attributes of the event, the driver, and the context to reduce the likelihood of bias. Hierarchical model structures offer the advantage of driver-level predictors to parameterize the effects of event attributes and contexts. The models thus reflect how driver decisions are executed: drivers with particular characteristics (one level) find themselves in contexts in which they execute specific driving maneuvers (second level), which lead to certain outcomes. Suggestions for further research include testing with additional data sets and potential applications to analysis of crash surrogates.

AB - Naturalistic driving studies have been conducted over the past 5 years or more and have commonly reviewed video and kinematic data to identify and analyze crash, near-crash, and critical-incident events. But statistical methods that are applicable to these event data are needed. This paper addresses two issues in model development for naturalistic driving event data: the test for omitted-variable bias and the exploration of the advantages of hierarchical model structures in data analysis. With roadway departure event data from the 100-Car Naturalistic Driving Study conducted at Virginia Tech Transportation Institute, Blacksburg, Virginia, logit models were used to estimate the probability that a crash or a near crash would occur, rather than a critical incident. The models indicated a substantial omitted-variable bias for estimation of the effect of context variables but little difference for driver variables. These tests indicated that modeling of naturalistic event data should have included variables that described the attributes of the event, the driver, and the context to reduce the likelihood of bias. Hierarchical model structures offer the advantage of driver-level predictors to parameterize the effects of event attributes and contexts. The models thus reflect how driver decisions are executed: drivers with particular characteristics (one level) find themselves in contexts in which they execute specific driving maneuvers (second level), which lead to certain outcomes. Suggestions for further research include testing with additional data sets and potential applications to analysis of crash surrogates.

UR - http://www.scopus.com/inward/record.url?scp=84855211714&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84855211714&partnerID=8YFLogxK

U2 - 10.3141/2236-06

DO - 10.3141/2236-06

M3 - Article

SP - 49

EP - 57

JO - Transportation Research Record

JF - Transportation Research Record

SN - 0361-1981

IS - 2236

ER -