Civil, Construction, and Environmental Engineering, 260 H.M. Comer Hall, Box 870205, The University of Alabama, Tuscaloosa, AL 35487-0205, United States.
Accid Anal Prev. 2011 Sep;43(5):1644-51. doi: 10.1016/j.aap.2011.03.018. Epub 2011 Apr 19.
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.
阿拉巴马州教育部和州长学校班车安全带研究小组授权并资助了一个研究项目，旨在调查阿拉巴马州校车上的安全带对学生的影响。本文对该研究的一个重要组成部分进行了实证分析——影响学生在校车上是否系安全带的因素。离散选择模型框架被应用于量化各种因素的相对影响。为了获取关于学生个体特征和出行特征的细粒度信息，开发了一种新的数据收集协议。调查了 11 个变量，其中 8 个变量具有显著影响。这 8 个变量分别是：年龄、性别、学生的家庭所在县、学生的乘车路程、一天中的时间、随车照管员的存在和积极参与情况，以及两级别的司机参与情况。所得到的模型很好地拟合了数据，并揭示了文献中被忽视或低估的几个趋势。该模型还可以用于预测由于影响因素的变化而导致的安全带使用率的变化。这有助于确定提高合规率的最具成本效益的方法，而提高合规率对安全带带来的附加安全效益的实现至关重要。本文首次使用严格的统计建模技术来量化一系列变量的相对影响。这项研究将为文献做出贡献，并为学校交通管理实践提供有价值的见解。
