The before-after study is still the most popular method used by traffic engineers and transportation safety analysts for evaluating the effects of an intervention. However, this kind of study may be plagued by important methodological limitations, which could significantly alter the study outcome. They include the regression-to-the-mean (RTM) and site-selection effects. So far, most of the research on these biases has focused on the RTM. Hence, the primary objective of this study consists of presenting a method that can reduce the site-selection bias when an entry criterion is used in before-after studies for continuous (e.g. speed, reaction times, etc.) and count data (e.g. number of crashes, number of fatalities, etc.). The proposed method documented in this research provides a way to adjust the Naive estimator by using the sample data and without relying on the data collected from the control group, since finding enough appropriate sites for the control group is much harder in traffic-safety analyses. In this study, the proposed method, a.k.a. Adjusted method, was compared to commonly used methods in before-after studies. The study results showed that among all methods evaluated, the Naive is the most significantly affected by the selection bias. Using the CG, the ANCOVA, or the EB method based on a control group (EBCG) method can eliminate the site-selection bias, as long as the characteristics of the control group are exactly the same as those for the treatment group. However, control group data that have same characteristics based on a truncated distribution or sample may not be available in practice. Moreover, site-selection bias generated by using a dissimilar control group might be even higher than with using the Naive method. The Adjusted method can partially eliminate site-selection bias even when biased estimators of the mean, variance, and correlation coefficient of a truncated normal distribution are used or are not known with certainty. In addition, three actual datasets were used to evaluate the accuracy of the Adjusted method for estimating site-selection biases for various types of data that have different mean and sample-size values.

This research explores the red light running phenomena and offer a better understanding of the factors associated with it. The red light running is a type of traffic violation that can lead to angle crash and the most common counter measure is installing a red light running cameras. Red light running cameras some time can reduce the rates of red light running but because of the increased worry of the public towards crossing the intersection it can cause an increase in rear end crashes. Also the public opinion of the red light running cameras is that they are a revenue generator for the local counties and not a concern of public safety. Further more, they consider this type of enforcement as violation of privacy. There was two ways to collect the data needed for the research. One way is through a tripod cameras setup temporarily placed at the intersection. This setup can collect individual vehicles caught in the change phase with specific information about their reactions and conditions. This required extensive manual analysis for the recorded videos plus data could not be collected during adverse weather conditions. The second way was using traffic monitoring cameras permanently located at the site to collect red light running information and the simultaneous traffic conditions. This system offered more extensive information since the cameras monitor the traffic 24/7 collecting data directly. On the other hand this system lacked the ability to identify the circumstances associated with individual red light running incidents. The research team finally decided to use the two methods to study the red light running phenomena aiming to combine the benefits of the two systems. During the research the team conducted an experiment to test a red light running countermeasure in the field and evaluate the public reaction and usage of this countermeasure. The marking was previously tested in a driving simulator and proved to be successful in helping the drivers make better stop/go decisions thus reducing red light running rates without increasing the rear-end crashes. The experiment was divided into three phases; before marking installation called "before", after marking installation called "after", and following a media campaign designed to inform the public about the use of the marking the third phase called "after media" The behavior study that aimed at analyzing the motorist reactions toward the signal change interval identified factors which contributed to red light running. There important factors were: distance from the stop bar, speed of traffic, leading or following in the traffic, vehicle type. It was found that a driver is more likely to run red light following another vehicle in the intersection. Also the speeding vehicles can clear the intersection faster thus got less involved in red light running violations. The proposed "Signal Ahead" marking was found to have a very good potential as a red light running counter measure. The red light running rates in the test intersection dropped from 53 RLR/hr/1000veh for the "before" phase, to 24 RLR/hr/1000veh for the "after media" phase. The marking after media analysis period found that the marking can help the driver make stop/go decision as the dilemma zone decreased by 50 ft between the "before" and the "after media" periods. Analysis of the traffic condition associated with the red light running it revealed that relation between the traffic conditions and the red light running is non-linear, with some interactions between factors. The most important factors included in the model were: traffic volume, average speed of traffic, the percentage of green time, the percentage of heavy vehicles, the interaction between traffic volume and percentage of heavy vehicles. The most interesting finding was the interaction between the volume and the percent of heavy vehicles. As the volume increased the effect of the heavy vehicles reversed from reducing the red light running to increasing the red light. This finding may be attributed to the sight blocking that happens when a driver of a passenger car follows a larger heavy vehicle, and can be also explained by the potential frustration experienced by the motorist resulting from driving behind a bigger vehicle.