Ordered Probability Models Research Articles

ICT Use, Environmental Quality Perception and Farmers’ Participation in Domestic Waste Separation: Micro-Survey Data from China

Encouraging farmers to participate in domestic waste sorting is an important initiative to optimize rural habitats and build a beautiful countryside. Using data from a sample of 2126 farmers obtained from a Chinese micro-survey, this paper empirically investigates the impact of ICT use on farmers’ domestic waste classification using OLS and ordered probability models and examines the mediating role of environmental quality perception. The study shows that ICT use has a significant negative effect on farmers’ environmental quality perceptions and a significant positive effect on farmers’ domestic waste sorting. Furthermore, the direct positive effect of ICT use on farmers’ domestic waste sorting is greater than its negative indirect effect through environmental quality perceptions. Finally, farmers with ICT use are more willing to participate in domestic waste sorting. This suggests that farmers may have a tendency to complain and express dissatisfaction on the internet but still behave in a way that is participatory in waste sorting. The results of the study still hold after a rigorous robustness test. In addition, there are significant differences in the impact of ICT use on different age and income groups, so policies should be tailored to different groups. More attention should be paid to the environmental welfare effects on older and lower-income groups.

Trajectory-based conflict investigations involving two-wheelers and cars at non-signalized intersections with computer vision

Unsafe acts at non-signalized intersections have become a primary contributor to traffic accidents and fatalities. Whereas many studies have focused on non-signalized intersections in the past decade, capturing and recording road users’ micro-behavior and risk in the mixed traffic flow remain challenging. A large number of two-wheelers (e.g., bicycles and e-bikes) appear at non-signalized intersections, in which the conflict behaviors are highly unpredictable. In this study, conflicts involving two-wheelers and cars at non-signalized intersections were investigated based on trajectory data collected with a YOLOv3-based framework automatically. A novel conflict identification algorithm was developed to gather and process microscopic trajectory data. To detect conflict behaviors involving two-wheelers and cars, near-crash identification was employed with a post-encroachment time indicator that also contributes to demonstrating the effect of vehicle order on conflict severity from an unprecedented perspective. The proposed framework was applied to a case study at a university campus in Shanghai. To explore the relationship between contributing factors and conflict severity, a significance test and ordered probability models were implemented using 10,304 conflicts collected from video data. The statistical analysis disclosed that conflicts involving e-bikes accounted for the highest proportion, and the order of vehicles (e.g., pre-encroachment vehicle and post-encroachment vehicle) has different effects on conflict severity. The analytical results with risk assessment can contribute to developing intersection-specific countermeasures for traffic safety from the perspectives of education, engineering, and law enforcement. The trajectory-based framework can be adapted to intelligent transportation systems to enhance safety management.

A correlated random parameters with heterogeneity in means approach of deer-vehicle collisions and resulting injury-severities

This paper investigates deer-vehicle collisions and their resulting injury-severities in a threefold approach. Two random parameters binary logit models with heterogeneity in means are estimated to analyze: (a) the likelihood of witnessing deer on the roadway; and (b) the likelihood of a vehicle hitting deer on the roadway. Additionally, a novel variant of ordered probability models, namely the correlated random parameters ordered logit model with heterogeneity in means, is estimated to explore the factors contributing to different driver injury-severity outcomes resulting from deer-vehicle collisions. A database of crashes maintained by the Pennsylvania Department of Transportation for the year 2018 is used for the analysis. The findings reveal that it is more likely to witness deer in rural locations, in dark lighting conditions, and during deer breeding season – when deer are more active in moving from one location to another. In addition, it is more likely for vehicles to hit deer in roads with speed limits above 55 mph, and during the deer breeding season. Furthermore, female drivers are found to be more likely to hit deer as compared to male drivers. Finally, airbag deployment and post-crash overturning of the vehicle are both associated with major injuries and fatalities; whereas, use of restraints is found to prevent injuries or fatalities. Employing the correlated random parameters and the heterogeneity in means approaches offers additional insights about the effects of unobserved heterogeneity on factors contributing to the likelihood of witnessing and hitting deer on the roadway, as well as the driver injury-severity levels resulting from deer-vehicle collisions.

The joint effect of weather and lighting conditions on injury severities of single-vehicle accidents

This study seeks to identify and analyze variations in the effect of contributing factors on injury severities of single-vehicle accidents across various lighting and weather conditions. To that end, injury-severity data from single-vehicle, injury accidents occurred in Scotland, United Kingdom in 2016 and 2017 are statistically modeled. Upon the conduct of likelihood ratio tests, separate models of accident injury severities are estimated for various combinations of weather and lighting conditions taking also into account the presence and operation of roadside lighting infrastructure. To account for the possibility of unobserved regimes underpinning the injury-severity mechanism, the zero-inflated hierarchical ordered probit approach with correlated disturbances is employed. This approach also relaxes the fixed threshold restriction of the traditional ordered probability models and captures systematic unobserved variations between the underlying regimes. The model estimation results show that a wide range of accident, vehicle, driver, trip and location characteristics have varying impacts on injury severities when different weather and lighting conditions are jointly considered. Even though several factors are identified to have overall consistent effects on injury severities, the simultaneous impact of unfavorable weather and lighting conditions is found to introduce significant variations, especially in the effect of vehicle- and driver-specific characteristics. The findings of this study can be leveraged in vehicle-to-infrastructure or in-vehicle communication technologies that can assist drivers in their responses against hazardous environmental conditions.

Bicycle Level of Service Model for the Cycloruta, Bogota, Colombia

Abstract Segment videos were produced at different peaks to reflect different sampling criteria like land use characteristics, trails, Ciclocarrils and Ciclovia. Each segment was filmed for 20–40 seconds during bicycle rides at a speed of about 5km/h with a camera strapped, at an angle of 45 degrees, on the head. Curb lane variables such as bicycle pathway widths, curb lane motorised volume (veh/h) and vehicle speed (km/h), bicycle volume on segment, and median width were recorded in addition to secondary data. About 1,360 ratings were acquired from study participants and used in the estimation process. Ordered probability models were used to estimate random parameters of cyclists LOS perception to account for unobserved heterogeneity for all respondents. The deviance (1.085) and Pearson Chi-Square (2.309) with 1,635 degree of freedom at 0.05 level of significance shows that our model provides a better fit of the data. The study observed that BLOS was strongly influenced by side path separation, vehicle speed, motorised traffic volume and conflicts with pedestrians. However, many other factors were found to have high probabilities to influence level of service with unit change. They include bicycle lane width, wide outside lane, pavement conditions, trees and benches, daylight, gender and experience of cyclist. The impact of the variety of observed factors affecting bicyclists reveal the nature and character of urban transportation in Bogota which suggests a range of important trade-offs in further planning and management of the Cicloruta bicycle paths.

Comparison of the linear regression, multinomial logit, and ordered probability models for predicting the distribution of thermal sensation

This study compares the linear regression model, ordered probability model, and multinomial logit model for prediction of the individual thermal sensation votes (TSVs) and TSV distributions under given conditions. Two thermal comfort datasets were used to develop and evaluate the models. One dataset was taken from an indoor thermal comfort survey conducted in Pakistan, and the other was taken from an outdoor thermal comfort survey conducted in Tianjin, China. The data were divided into training and validation datasets. The training datasets were used for model development. The developed models were then used to predict new cases in the validation dataset. The predictive capability of the three models were systematically evaluated and compared to examine how well the developed models predicted individual TSVs and TSV distributions for the validation dataset. The results showed that the ordered probability model and the multinomial logit model correctly predicted around 50% of the individual TSVs, whereas the accuracy of the linear regression model was only around 20 to 40%. In addition, the chi-square statistics show that the ordered probability model and the multinomial logit model better predicted the TSV distributions than the linear regression model.

Contributory fault and level of personal injury to drivers involved in head-on collisions: Application of copula-based bivariate ordinal models

The main objective of this study is to simultaneously investigate the degree of injury severity sustained by drivers involved in head-on collisions with respect to fault status designation. This is complicated to answer due to many issues, one of which is the potential presence of correlation between injury outcomes of drivers involved in the same head-on collision. To address this concern, we present seemingly unrelated bivariate ordered response models by analyzing the joint injury severity probability distribution of at-fault and not-at-fault drivers. Moreover, the assumption of bivariate normality of residuals and the linear form of stochastic dependence implied by such models may be unduly restrictive. To test this, Archimedean copula structures and normal mixture marginals are integrated into the joint estimation framework, which can characterize complex forms of stochastic dependencies and non-normality in residual terms. The models are estimated using 2013 Virginia police reported two-vehicle head-on collision data, where exactly one driver is at-fault. The results suggest that both at-fault and not-at-fault drivers sustained serious/fatal injuries in 8% of crashes, whereas, in 4% of the cases, the not-at-fault driver sustained a serious/fatal injury with no injury to the at-fault driver at all. Furthermore, if the at-fault driver is fatigued, apparently asleep, or has been drinking the not-at-fault driver is more likely to sustain a severe/fatal injury, controlling for other factors and potential correlations between the injury outcomes. While not-at-fault vehicle speed affects injury severity of at-fault driver, the effect is smaller than the effect of at-fault vehicle speed on at-fault injury outcome. Contrarily, and importantly, the effect of at-fault vehicle speed on injury severity of not-at-fault driver is almost equal to the effect of not-at-fault vehicle speed on injury outcome of not-at-fault driver. Compared to traditional ordered probability models, the study provides evidence that copula based bivariate models can provide more reliable estimates and richer insights. Practical implications of the results are discussed.

A random thresholds random parameters hierarchical ordered probit analysis of highway accident injury-severities

This study uses highway accident data collected in the State of Washington, between 2011 and 2013, to study the factors that affect accident injury-severities. To account for the fixed thresholds limitation of the traditional ordered probability models – which typically leads to incorrect estimation of outcome probabilities for the intermediate categories – and for the possibility of unobserved factors systematically varying across the observations, a random thresholds hierarchical ordered probit model with random parameters is estimated. This approach simultaneously allows the explanatory parameters to vary across roadway segments, and the thresholds to vary both as a function of explanatory parameters and across the observations, thus accounting for unobserved and threshold heterogeneity, respectively. Using goodness-of-fit measures, likelihood ratio tests and forecasting accuracy measures, the model estimation results are compared with the hierarchical and fixed thresholds ordered probit model counterparts, with fixed and random parameters. The comparative assessment among the ordered probit modeling approaches reveals the relative benefits and the overall statistical superiority of the random thresholds random parameters hierarchical ordered probit model.

Obesity and the Demand for Canadian Physician Services

The objective of this study is to determine the role that obesity plays in how often Canadians visit their family doctors or general practitioners. Doctor visits are analyzed using mixtures of ordered probability models applied to sample survey data from the 2010 Canadian Community Health Survey. This procedure is shown to be superior in terms of likelihood criteria to the more usual one involving count models of doctor visits. The main result is that obesity is one of the leading causes of doctor visits. Obesity has become more important in the demand for physician services than smoking for all Canadians. Other factors including diabetes, the individual’s level of education, position in the income distribution, and drinking behavior are also important. The application of latent class’s ordered probability models by age-group and gender leads to results which are different from what others have found. While obesity is shown to be a serious problem in Canada, it has not yet reached the stage which some researchers have described as critical.

Statistical analysis of pedestrian perceptions of sidewalk level of service in the presence of bicycles

Gaining a better understanding of pedestrian perceptions of level of service (LOS) in urban areas can be useful in developing strategies for providing a comfortable and safe walking environment. Using a sample of 114 respondents, this research studies how pedestrians perceive LOS on sidewalks shared with bicycles under various urban-street conditions. To do this, 15 carefully constructed video clips of Chinese urban sidewalks are presented to respondents and, after each 60-s video clip, respondents were asked to assess the pedestrian level of service in the video on a scale from LOS A (the most comfortable pedestrian environment) to LOS F (the least comfortable pedestrian environment). By estimating random parameters ordered probability models of respondents’ LOS assessments (to account for unobserved heterogeneity across respondents) we found, as expected, that pedestrian perceptions of LOS are strongly influenced by the pedestrian flow rate. However, many other factors were found to significantly affect LOS perceptions including sidewalk width, the presence of a barrier separating the sidewalk from motor-vehicle traffic, the presence of parking next to the sidewalk, the presence of businesses along the sidewalk, the bicycle flow rate, the speed of bicyclist, whether or not bicycles were riding against the flow of pedestrians, weather conditions, time of day, and age of the respondent. The specific impacts of the wide variety of factors that affect pedestrian perceptions of LOS suggest that a number of important tradeoffs need to be considered when planning pedestrian facilities.

Customer Satisfaction of Mobile-Internet Users: An Empirical Approximation for the Case of Spain

In this paper we study the mobile Internet services in Spain. We are especially interested in overall consumer satisfaction as well as with some of its attributes. We use micro data from a survey conducted by the Spanish Center for Sociological Research (CIS 2009). We have elaborated a model to analyze to what extent different service providers generate different levels of satisfaction of individuals when we control for other relevant variables. In this regard, we found statistically significant differences in favor of Vodafone for both overall satisfaction and for satisfaction with any other of its attributes. However, there are no significant differences between the other two considered companies (Movistar and Orange). The same conclusions were obtained when the model was estimated by OLS as when using ordered probability models (probit or logit).Our results also suggest that the communication quality of the service and its price are the main attributes that affect the overall satisfaction of the consumer.

Student Evaluation of Teaching, Formulation of Grade Expectations, and Instructor Choice: Explorations with Random-Effects Ordered Probability Models

This paper explores new angles of the student evaluation of teaching (SET) debate. The SET ratings of 179 full-time professors of the University of Puerto Rico-Bayamon were tracked from 1998–1999 to 2003–2004. Information on the characteristics of professors, students, and courses was analyzed. Two principal findings are documented. First, students’ grade expectations formulation process is significantly influenced by their quality and by the mean and variance of their professors’ historical grade distribution. Second, students’ willingness to take additional courses with a particular professor is directly related to the optimism of their relative expected grades. Thus, professors might be able to increase enrollment in their courses by: (a) increasing the students’ expected grades and (b) promoting the academic environment that would transform students’ pessimistic expectations into optimistic ones.

Educational mobility in Canada: results from the 2001 general social survey

This research examines mobility in the Canadian educational system and how it has evolved over the last 50 years of the twentieth century. Mixtures of ordered probability models which control for unobserved heterogeneity are estimated for both male and female respondents for two age groups: those between age 25 and 49 and those aged 50 or older in 2001. The main result is that educational attainments depend significantly less on family background variables for the younger cohorts. This is a qualitatively different result from what Canadian sociologists have found. They argue that there is little mobility in the Canadian educational system which has remained much as it was at the beginning of the twentieth century. The data used in the analysis comes from the 2001 general social survey.

Analysis of left-turn crash injury severity by conflicting pattern using partial proportional odds models

The purpose of this study is to examine left-turn crash injury severity. Left-turning traffic colliding with opposing through traffic and with near-side through traffic are the two most frequently occurring conflicting patterns among left-turn crashes (Patterns 5 and 8 in the paper, respectively), and they are prone to be severe. Ordered probability models with either logit or probit function is commonly applied in crash injury severity analyses; however, its critical assumption that the slope coefficients do not vary over different alternatives except the cut-off points is usually too restrictive. Partial proportional odds models are generalizations of ordered probability models, for which some of the beta coefficients can differ across alternatives, were applied to investigate Patterns 5 and 8, and the total left-turn crash injuries. The results show that partial proportional odds models consistently perform better than ordered probability models. By focusing on specific conflicting patterns, locating crashes to the exact crash sites and relating approach variables to crash injury in the analysis, researchers are able to investigate how these variables affect left-turn crash injuries. For example, opposing through traffic and near-side crossing through traffic in the hour of collision were identified significant for Patterns 5 and 8 crash injuries, respectively. Protected left-turn phasing is significantly correlated with Pattern 5 crash injury. Many other variables in driver attributes, vehicular characteristics, roadway geometry design, environmental factors, and crash characteristics were identified. Specifically, the use of the partial proportional formulation allows a much better identification of the increasing effect of alcohol and/or drug use on crash injury severity, which previously was masked using the conventional ordered probability models.

ESTIMATING CONSUMER SATISFACTION: OLS VERSUS ORDERED PROBABILITY MODELS

This paper reviews the use of logit and probit models in marketing and focuses on demonstrating the use of ordered probability models. This type of model is appropriate for many applications in marketing and business where the dependent variable of interest is ordinal (e.g., likert scales). A comparison between the properties of the ordinary least squares (OLS) model and ordered logit and probit models is made using consumer satisfaction data on automobiles. This comparison between the two models shows that the use of OLS for ordered categorical data gives misleading results and produces biased estimates, leading to inaccurate hypothesis testing. The paper concludes that ordered probability models, such as the ones illustrated, should be employed in marketing and business research where the dependent variable is ordinal.