Abstract
Driver inattention has long been an issue for traffic safety advocates. Drivers may only briefly look away to change the radio station, answer a phone call, send a text message, or speak to a passenger; however, taking their eyes off the road decreases driver awareness and increases the likelihood of a collision. One of the main sources of driver inattention is the use of cell phones while driving. In California, lawmakers sought to address the dangers of cell phone use while driving by banning the use of hand-held cell phones while operating a vehicle. California’s ban went into effect July 1, 2008, and while there have been multiple changes to the law in the five years since, there has been little evaluation as to whether the law achieved its goal of reducing accidents. Using accident report data compiled in the Statewide Integrated Traffic Records System, I performed three regression analyses to determine whether California’s law prohibiting cell phone use while driving resulted in fewer traffic accidents in the year after the law went into effect compared to the year prior. The first two logistic regression models measure fatal accidents and injury accidents for both years to establish the impact of cell phone use on these types of accidents. The third model measures the impact of the law on accidents involving cell phone use. The general causal factors identified are the year during which the accident occurred, driver behavior, driver demographics, accident time, accident location, weather conditions, and road conditions. In evaluating the final regression results for Model 1 (Accident Involving Fatality=1), the key explanatory variable (Cell Phone in Use) was not statistically significant. For Model 2 (Accident with Injury=1), Cell Phone in Use was both statistically significant and had a positive impact on the likelihood of being in an injury accident. Based on the results, a driver was 30.61 percent more likely to be involved in an accident involving an injury versus an accident with no injury or a fatality, while using a cell phone than a driver not using a cell phone, all else held constant. For Model 3 (Cell Phone Use While Driving and Being Involved in Accident=1), the results for the key explanatory variable (FY 2008/2009) were statistically significant and indicated that a driver was 42.79 percent less likely to be involved in an accident involving cell phone use than not involving a cell phone, in the year after the law went into effect compared to the prior year. These results must come with the caveat that not all factors influencing a driver’s behavior may be accounted for and that not all accidents resulting from driver cell phone use may be identified in the data set given the low Pseudo R2 values for each of the three models (Model 1 - 0.0631; Model 2 - 0.0248; Model 3 - 0.0540). Primarily, I recommended that a better method of data collection be identified to ensure the accuracy of conclusion drawn from data analysis. Possible suggestions include the development of best practices for law enforcement in identifying cell phone use at an accident and making the indication of cell phone use mandatory on the accident report. Secondary recommendations include using the demographic results of this study to inform public awareness campaigns to target those drivers most likely to be involved in an accident involving cell phone use and to inform driver education training.