Refined-scale panel data crash rate analysis using random-effects tobit model

Random effects tobit models are developed in predicting hourly crash rates with refined-scale panel data structure in both temporal and spatial domains. The proposed models address left-censoring effects of crash rates data while accounting for unobserved heterogeneity across groups and serial correlations within group in the meantime. The utilization of panel data in both refined temporal and spatial scales (hourly record and 1-mile roadway segments on average) exhibits strong potential on capturing the nature of time-varying and spatially varying contributing variables that is usually ignored in traditional aggregated traffic accident modeling. 1-year accident data and detailed traffic, environment, road geometry and surface condition data from a segment of I-25 in Colorado are adopted to demonstrate the proposed methodology. To better understand significantly different characteristics of crashes, two separate models, one for daytime and another for nighttime, have been developed. The results show major difference in contributing factors towards crash rate between daytime and nighttime models, implying considerable needs to investigate daytime and nighttime crashes separately using refined-scale data. After the models are developed, a comprehensive review of various contributing factors is made, followed by discussions on some interesting findings.     

Motivating business towards innovation: A panel data study using dynamic capability framework

National strategic dynamic capabilities refer to the nations' ability to acquire, develop, and transfer national resources into innovation. This would help provoke firms and individuals’ capabilities to sense and seize to expedite innovative performance transformation. This study suggests that through the dynamic capabilities (DC) framework, the government can increase Innovation by channelizing the economic innovation system indicators. Macroeconomic indicators from 62 countries were taken to assess the significance of indicator-based dynamic capabilities model. The results confirm that sensing and seizing indices motivate Innovation, while the cost of business negatively moderates sensing. This model suggests that governments can motivate Innovation by exploring the indicators which are determining the sensing and seizing environment of decision-makers in businesses.     

An alternative approach to trend analysis in accident data

Traditional statistical theory is the most common tool used for trend analysis in accident data. In this paper, we point at some serious problems in using this theory in a practical safety management setting. An alternative approach is presented and discussed in which focus is on observable quantities and expressing uncertainties regarding these rather than on hypothetical probability distributions.     

Governance mechanisms for sharing of health data: An approach towards selecting attributes for complex discrete choice experiment studies

BackgroundDiscrete Choice Experiment (DCE) is a well-established technique to elicit individual preferences, but it has rarely been used to elicit governance preferences for health data sharing.ObjectivesThe aim of this article was to describe the process of identifying attributes for a DCE study aiming to elicit preferences of citizens in Sweden, Iceland and the UK for governance mechanisms for digitally sharing different kinds of health data in different contexts.MethodsA three-step approach was utilised to inform the attribute and level selection: 1) Attribute identification, 2) Attribute development and 3) Attribute refinement. First, we developed an initial set of potential attributes from a literature review and a workshop with experts. To further develop attributes, focus group discussions with citizens (n = 13), ranking exercises among focus group participants (n = 48) and expert interviews (n = 18) were performed. Thereafter, attributes were refined using group discussion (n = 3) with experts as well as cognitive interviews with citizens (n = 11).ResultsThe results led to the selection of seven attributes for further development: 1) level of identification, 2) the purpose of data use, 3) type of information, 4) consent, 5) new data user, 6) collector and 7) the oversight of data sharing. Differences were found between countries regarding the order of top three attributes. The process outlined participants’ conceptualisation of the chosen attributes, and what we learned for our attribute development phase.ConclusionsThis study demonstrates a process for selection of attributes for a (multi-country) DCE involving three stages: Attribute identification, Attribute development and Attribute refinement. This study can contribute to improve the ethical aspects and good practice of this phase in DCE studies. Specifically, it can contribute to the development of governance mechanisms in the digital world, where people's health data are shared for multiple purposes.     

Novice riders and the predictors of riding without motorcycle protective clothing

Background

While helmet usage is often mandated, few motorcycle and scooter riders make full use of protection for the rest of the body. Little is known about the factors associated with riders’ usage or non-usage of protective clothing.

Methods

Novice riders were surveyed prior to their provisional licence test in NSW, Australia. Questions related to usage and beliefs about protective clothing, riding experience and exposure, risk taking and demographic details. Multivariable Poisson regression models were used to identify factors associated with two measures of usage, comparing those who sometimes vs rarely/never rode unprotected and who usually wore non-motorcycle pants vs motorcycle pants.

Results

Ninety-four percent of eligible riders participated and usable data was obtained from 66% (n = 776). Factors significantly associated with riding unprotected were: youth (17–25 years) (RR = 2.00, 95% CI: 1.50–2.65), not seeking protective clothing information (RR = 1.29, 95% CI = 1.07–1.56), non-usage in hot weather (RR = 3.01, 95% CI: 2.38–3.82), awareness of social pressure to wear more protection (RR = 1.48, 95% CI: 1.12–1.95), scepticism about protective benefits (RR = 2.00, 95% CI: 1.22–3.28) and riding a scooter vs any type of motorcycle. A similar cluster of factors including youth (RR = 1.17, 95% CI: 1.04–1.32), social pressure (RR = 1.32, 95% CI: 1.16–1.50), hot weather (RR = 1.30, 95% CI: 1.19–1.41) and scooter vs motorcycles were also associated with wearing non-motorcycle pants. There was no evidence of an association between use of protective clothing and other indicators of risk taking behaviour.

Conclusions

Factors strongly associated with non-use of protective clothing include not having sought information about protective clothing and not believing in its injury reduction value. Interventions to increase use may therefore need to focus on development of credible information sources about crash risk and the benefits of protective clothing. Further work is required to develop motorcycle protective clothing suitable for hot climates.     

A mixed logit analysis of two-vehicle crash severities involving a motorcycle

Using motorcycle crash data for Iowa from 2001 to 2008, this paper estimates a mixed logit model to investigate the factors that affect crash severity outcomes in a collision between a motorcycle and another vehicle. These include crash-specific factors (such as manner of collision, motorcycle rider and non-motorcycle driver and vehicle actions), roadway and environmental conditions, location and time, motorcycle rider and non-motorcycle driver and vehicle attributes. The methodological approach allows the parameters to vary across observations as opposed to a single parameter representing all observations. Our results showed non-uniform effects of rear-end collisions on minor injury crashes, as well as of the roadway speed limit greater or equal to 55 mph, the type of area (urban), the riding season (summer) and motorcyclist's gender on low severity crashes. We also found significant effects of the roadway surface condition, clear vision (not obscured by moving vehicles, trees, buildings, or other), light conditions, speed limit, and helmet use on severe injury outcomes.     

A mixed-Weibull regression model for the analysis of automotive warranty data

This paper presents a case study regarding the reliability analysis of some automotive components based on field failure warranty data. The components exhibit two different failure modes, namely early and wearout failures, and are mounted on different vehicles, which differ among themselves for car model and engine type, thus involving different operating conditions. Hence, the failure time of each component is a random variable with a bimodal pdf which also depends upon a vector of covariates that indexes the specific operating condition. Then, a mixed-Weibull distribution, where the pdf of each subpopulation (namely the ‘weak’ and ‘strong’ subpopulation) depends on the covariates through the scale parameter, is used to analyze the component lifetime. A Fortran algorithm for the maximum likelihood estimation of model parameters has been implemented and a stepwise procedure, in its backwards version, has been used to test the significance of covariates and to construct the regression model. The presence of a weak subpopulation has been verified and the fraction of weak units in the population has also been estimated. Finally, the adequacy of the proposed model to fit the observed data has been assessed.     

Factors influencing students' usage of school bus seat belts: an empirical analysis of the Alabama pilot project

The Alabama State Department of Education and the Governor's Study Group on School Bus Seat Belts authorized and funded a research project to investigate the effects of lap-shoulder seat belts on Alabama school buses. This article performs an empirical analysis to address an important component of the study – factors that impact students’ decisions about wearing seat belts or not on school buses. Discrete choice modeling framework is applied to quantify relative influences of various factors. To obtain the disaggregate level information on individual student's characteristics and trip properties, a new data collection protocol is developed. Eleven variables are investigated and eight of them are found to have significant impacts. They are age, gender, the home county of a student, a student's trip length, time of day, presence and active involvement of bus aide, and two levels of bus driver involvement. The resulting model fits the data well and reveals several trends that have been overlooked or underestimated in the literature. The model can also be used to predict the change of seat belt usage rate caused by the change of impact factors. This is helpful in identifying the most cost-effective ways to improve compliance rate, which is critical to bring the added safety benefit of seat belts into effect. This article is the first to quantify relative impacts of a range of variables using rigorous statistical modeling techniques. This study will contribute to the literature and provide valuable insights to the practice of school transportation management.     

Panel data analysis—advantages and challenges

We explain the proliferation of panel data studies in terms of (i) data availability, (ii) the more heightened capacity for modeling the complexity of human behavior than a single cross-section or time series data can possibly allow, and (iii) challenging methodology. Advantages and issues of panel data modeling are also discussed. This invited paper is discussed in the comments available at: , , , , , , , .     

Internet use and insurance growth: evidence from a panel of OECD countries

This empirical study aimed to investigate the effect of internet use on the demand for insurance using a sample of OECD countries during the period from 2007 to 2017. We employed a battery of panel data models, taking into account various unusual patterns, such as heterogeneity. This study contributes to the technological innovations and communication literature at different stages of the provision and development of life and non-life insurance activities. The empirical results show that internet use positively affects non-life insurance activities. This finding is well recognised by several estimated static and dynamic panel models. Conversely, the empirical investigation revealed no global impact of internet use on the demand for life insurance. Lastly, an assessment of traditional life insurance models demonstrated significant impacts of several socio-economic variables on the life insurance line of business.     

Chunking: A procedure to improve naturalistic data analysis

Every year, traffic accidents are responsible for more than 1,000,000 fatalities worldwide. Understanding the causes of traffic accidents and increasing safety on the road are priority issues for both legislators and the automotive industry. Recently, in Europe, the US and Japan, significant public funding has been allocated for performing large-scale naturalistic driving studies to better understand accident causation and the impact of safety systems on traffic safety. The data provided by these naturalistic driving studies has never been available before in this quantity and comprehensiveness and it promises to support a wide variety of data analyses. The volume and variety of the data also pose substantial challenges that demand new data reduction and analysis techniques. This paper presents a general procedure for the analysis of naturalistic driving data called chunking that can support many of these analyses by increasing their robustness and sensitivity. Chunking divides data into equivalent, elementary chunks of data to facilitate a robust and consistent calculation of parameters. This procedure was applied, as an example, to naturalistic driving data from the SeMiFOT study in Sweden and compared with alternative procedures from past studies in order to show its advantages and rationale in a specific example. Our results show how to apply the chunking procedure and how chunking can help avoid bias from data segments with heterogeneous durations (typically obtained from SQL queries). Finally, this paper shows how chunking can increase the robustness of parameter calculation, statistical sensitivity, and create a solid basis for further data analyses.     

A review of risk factors and patterns of motorcycle injuries

Per vehicle mile traveled, motorcycle riders have a 34-fold higher risk of death in a crash than people driving other types of motor vehicles. While lower-extremity injuries most commonly occur in all motorcycle crashes, head injuries are most frequent in fatal crashes. Helmets and helmet use laws have been shown to be effective in reducing head injuries and deaths from motorcycle crashes. Alcohol is the major contributing factor to fatal crashes. Enforcement of legal limits on the blood alcohol concentration is effective in reducing motorcycle deaths, while some alcohol-related interventions such as a minimal legal drinking age, increased alcohol excise taxes, and responsible beverage service specifically for motorcycle riders have not been examined. Other modifiable protective or risk factors comprise inexperience and driver training, conspicuity and daytime headlight laws, motorcycle licensure and ownership, riding speed, and risk-taking behaviors. Features of motorcycle use and potentially effective prevention programs for motorcycle crash injuries in developing countries are discussed. Finally, recommendations for future motorcycle-injury research are made.     

A hierarchical modeling approach to accelerated degradation testing data analysis: A case study

In this case study, we investigate the degradation process of light‐emitting diodes (LEDs), which is used as a light source in DNA sequencing machines. Accelerated degradation tests are applied by varying temperature and forward current, and the light outputs are measured by a computerized measuring system. A degradation path model, which connects to the LED function recommended in Mitsuo (1991), is used in describing the degradation process. We consider variations in both measurement errors and degradation paths among individual test units. It is demonstrated that the hierarchical modeling approach is flexible and powerful in modeling a complex degradation process with nonlinear function and random coefficient. After fitting the model by maximum likelihood estimation, the failure time distribution can be obtained by simulation. Copyright © 2010 John Wiley & Sons, Ltd.     

The impact of outlet densities on alcohol-related crashes: a spatial panel approach

BACKGROUND: A number of studies using cross-sectional data have demonstrated that the availability of alcohol, measured by the number and types of alcohol outlets, is directly related to numerous measures associated with drinking and driving. The current study contributes the first observation of relationships over time between alcohol outlet densities on one hand and both automobile crashes and related injuries on the other hand. METHOD: The study examined longitudinal data from 581 consistently defined zip code areas represented in the California Index Locations Database, a geographic information system that coordinates population and ecological data with spatial attributes for areas across the state. Six years of data were collected on features of local populations (e.g., demographics, household size) and places (e.g., retail markets) thought to be related to two measures of automobile crashes (hospital discharges related to car crash injuries geocoded to the zip code of patient residence, and police reports associated with car crashes geocoded to the zip code of crash location). Both crash measures were positively associated with two outlet types: bars, and off-premise outlets. Additionally, restaurants appear to provide a protective effect relative to the residence-based measure. Crash rates were also related to changes in population and place characteristics using random effects models with controls for spatial autocorrelation (nxt=3486 observations). Changes in population and place characteristics of adjacent (spatially lagged) areas were also considered. RESULTS: Over time, both local and lagged population and place characteristics were related to automobile crash-related measures. CONCLUSION: Controlling for cross-sectional differences between zip code areas, changes in numbers of licensed alcohol retail establishments, especially bars and off-premise outlets, affect rates of car crashes and related injuries.     

Investigation of seasonal variations of Beijing pavement condition data using unevenly spaced dynamic panel data model

Beijing municipal highway administration started to collect pavement condition data on its major expressways since 2006. It is advised in the Chinese practice standard that data collection shall be conducted on annual basis. However, pavement data are usually collected at different seasons of the year, which may cause significant seasonal variations in the observed condition. Moreover, for some reasons, data are missing for some of the pavement sections at certain years, which could bring up difficulties in performance model estimation and inference. These concerns have been simply neglected in past practice. This study proposed an unevenly spaced dynamic panel data model to investigate the seasonal patterns of a performance indicator called Ride Quality Index (RQI). A quasi-differencing approach was adopted for the estimation. Data collected from the 5th Ring Road of Beijing were used in the case study. It was found that RQI data collected during the fall season are expected to be lower than that collected during the spring or summer seasons. Findings from this research would be helpful to pavement engineers in using unevenly spaced pavement condition data for future condition estimation.     

Issues in motorcycle sensory and cognitive conspicuity: The impact of motorcycle low-beam headlights and riding experience on drivers’ decisions to turn across the path of a motorcycle

Background

Crashes involving a passenger car and a motorcycle, where the car is turning across the path of the motorcycle, are a major crash type of motorcycle riders. It has been proposed that the incidence of such crashes could be reduced through improvements in motorcycle conspicuity. Operation of low-beam headlights on motorcycles has been discussed as one approach for improving the “sensory conspicuity” of motorcycles during daylight hours, whilst previous experience as a rider may serve to heighten “cognitive conspicuity” through raised awareness of motorcyclists on our roads.

Method

Twenty-three experienced car drivers with no riding experience (“drivers”) and 20 experienced car drivers who were also motorcycle riders (“driver–riders”) completed a series of trials in a driving simulator where their task in each trial was to turn ahead of an oncoming vehicle if they felt that they had sufficient room to do so safely. A key manipulation across trials was whether the oncoming vehicle was a motorcycle with headlights on, or a motorcycle with headlights off. Time gap (short, medium, long) was also manipulated.

Results

Results indicate that, at time gaps defined in the current study as short, low-beam headlights may confer some benefit in gap acceptance by encouraging drivers to accept fewer gaps ahead of a motorcycle with headlights on than ahead of a motorcycle with headlights off. No statistically significant differences in gap acceptance between the headlight conditions were found at either the medium or long time gaps. Irrespective of time gap, driver–riders were found to adopt a more efficient turn strategy than drivers with no direct riding experience.

Conclusions

Overall, the present research provides support for the use of low-beam headlights and riding experience as tools through which to augment the sensory and cognitive conspicuity of motorcycles, respectively. It is proposed that further research aim to explore directly the precise mechanisms underlying the observed effects.     

The relationship between the travelling speed and motorcycle styles in urban settings: A case study in Belgrade

This report examines the difference in the distribution of the speeds of different motorcycle styles and the difference in the distribution of speeds of particular motorcycle styles and cars. The relationship between the speed of motorcycles that possess and those that do not possess vehicle registration plates was also explored. The speed was measured at six different locations on main roads in the city of Belgrade, Serbia. The study confirmed that, on average, motorcyclists drive faster than drivers of cars, but extreme speeding is recorded 2.3 times more often by motorcyclists than by car drivers. In this research, the styles of motorcycles were divided into three different groups according to their average speeds. The first group consists of sport motorcycles, which were faster than the other styles. The second group consists of scooter motorcycles, which were slower. The third group consists of conventional, touring, enduro, and chopper motorcycles with speeds that were statistically not significantly different. According to the differences of the mean speed of motorcyclists who use and do not use vehicle registration plates, the use of the registration plates can be considered a significant indicator of traffic safety. By classifying motorcycles in the three different groups, the issue of “generalizing” motorcyclists as a unique group is avoided and can be taken into consideration for future studies of motorcyclist safety.     

Socioeconomic differences and motorcycle injuries: age at risk and injury severity among young drivers. A Swedish nationwide cohort study

STUDY OBJECTIVE: The study examines whether there are socioeconomic differences among young motorcycle drivers (aged 16-25) involved in road-traffic injuries with regard to age and injury severity. DESIGN: Nationwide retrospective register-based cohort study. SETTING AND PARTICIPANTS: Subjects born in 1970-1972 were extracted from the Swedish Population and Housing Census of 1985 (n = 334,070). Individual records from the 1985 census were linked to police-reported data and hospital-based data for the period 1988-1995 on the basis of a search for each subject's first registered road-traffic injury as a motorcycle driver (n = 2034). Information on household socioeconomic group was taken from the Swedish census of 1985. Two categories of crash severity were analysed (minor injury and severe/fatal injury), based on assessments of the police and according to length of hospitalization. MAIN RESULTS: Incidence of motorcycle injury varies considerably according to age of driver, reaching a peak at the age of 17. The greatest differences in injury risk between socioeconomic groups are present when their members are aged 17-19. At the age of 18, subjects belonging to low socioeconomic positions run a risk of injury occurrence 2.5 times higher than those belonging to the highest socioeconomic category. Young drivers in lower socioeconomic groups have higher odds for both minor and severe injuries than their counterparts in the highest socioeconomic group, but there is no further increase for the latter. CONCLUSIONS: The study demonstrates how crucial the first years of driving are in relation to injury, and how wide the gap is in terms of socioeconomic differences at these ages, suggesting that this is the most appropriate time for intervention.     

Modeling school bus seat belt usage: Nested and mixed logit approaches

School bus seat belt usage has been of great interest to the school transportation community. Understanding factors that influence students’ decisions about wearing seat belts or not is important in determining the most cost-effective ways to improve belt usage rate, and thus the seat belt safety benefits. This paper presents a rigorous empirical analysis on data from Alabama School Bus Pilot Project using discrete choice modeling framework. In order to collect relevant information on individual student-trips, a new data collection protocol is adopted. Three choice alternatives are considered in the study: wearing, not wearing, and improperly wearing seat belts. A student's choice probabilities of these alternatives are modeled as functions of the student's characteristics and trip attributes. The coefficients of the variables in the functions are estimated first using standard multinomial logit model. Moreover, to account for potential correlations among the three choice alternatives and individual-level preference and response heterogeneity among users, nested and mixed logit models are employed in the investigation. Eight significant influence factors are identified by the final models. Their relative impacts are also quantified. The factors include age, gender and the home county of a student, a student's trip length, time of day, seat location, presence and active involvement of bus aide, and two levels of bus driver involvement. The impact of the seat location on students’ seat belt usage is revealed for the first time by this study. Both hypotheses that some of the choice alternatives are correlated and that individual-level heterogeneity exists are tested statistically significant. In view of this, the nested and the mixed logit model are recommended over the standard multinomial logit model to describe and predict students’ seat belt usage behaviors. The final nested logit model uncovers a correlation between improper wearing and not wearing, indicating there are some unknown or unobserved contributing factors that are common to these two choices. In the final random-parameter mixed logit model, individual preference heterogeneity is captured by random coefficients of county variables. Individual response heterogeneity is reflected in the random effect of a driver's remarks on students’ seat belt usage. Both recommended models are helpful in predicting seat belt usage rate quantitatively for given circumstances, and will provide valuable insights in practice of school transportation management.     

A comparison of the hazard perception ability of accident-involved and accident-free motorcycle riders

Hazard perception is the ability to read the road and is closely related to involvement in traffic accidents. It consists of both cognitive and behavioral components. Within the cognitive component, visual attention is an important function of driving whereas driving behavior, which represents the behavioral component, can affect the hazard perception of the driver. Motorcycle riders are the most vulnerable types of road user. The primary purpose of this study was to deepen our understanding of the correlation of different subtypes of visual attention and driving violation behaviors and their effect on hazard perception between accident-free and accident-involved motorcycle riders. Sixty-three accident-free and 46 accident-involved motorcycle riders undertook four neuropsychological tests of attention (Digit Vigilance Test, Color Trails Test-1, Color Trails Test-2, and Symbol Digit Modalities Test), filled out the Chinese Motorcycle Rider Driving Violation (CMRDV) Questionnaire, and viewed a road-user-based hazard situation with an eye-tracking system to record the response latencies to potentially dangerous traffic situations. The results showed that both the divided and selective attention of accident-involved motorcycle riders were significantly inferior to those of accident-free motorcycle riders, and that accident-involved riders exhibited significantly higher driving violation behaviors and took longer to identify hazardous situations compared to their accident-free counterparts. However, the results of the regression analysis showed that aggressive driving violation CMRDV score significantly predicted hazard perception and accident involvement of motorcycle riders. Given that all participants were mature and experienced motorcycle riders, the most plausible explanation for the differences between them is their driving style (influenced by an undesirable driving attitude), rather than skill deficits per se. The present study points to the importance of conceptualizing the influence of different driving behaviors so as to enrich our understanding of the role of human factors in road accidents and consequently develop effective countermeasures to prevent traffic accidents involving motorcycles.