Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
1The primary goals of SHRP 2 Safety Project S02, Integration of Analysis Methods and Devel- opment of Analysis Plan, were to (1) identify and prioritize key research questions related to the SHRP 2 Naturalistic Driving Study (NDS) and (2) determine important research elements (e.g., methods, data, questions) that analytical plans must address to explore these key ques- tions. The process began with more than 400 research issues defined by SHRP 2 Safety Projects S01, Development of Analysis Methods Using Recent Data, and S05, Design of the In-Vehicle Driv- ing Behavior and Crash Risk Study. These research issues were grouped into 27 global research questions, prioritized during Phase I of this project, and are further summarized in Appendix A (Boyle et al. 2009). The second phase of this project identified the highest-priority research questions accord- ing to the decision criteria developed in Phase I. These research questions and their rankings are presented in the main body of this report. They were developed to guide SHRP 2 and other researchers in preparing for analysis of the SHRP 2 NDS. They also were intended to guide the development of the statement of work for proposals for Project S08, Analysis of In-Vehicle Field Study Data and Countermeasure Implications. The project team has developed a recommended list of essential elements to be covered in the analysis project work plans. Recommendations are based on the teamâs expertise in this area and a detailed review of the analytical methods used by the S01 and S05 research teams. Example work plans are offered to illustrate the type of information that is relevant to a proposed work plan, as well as methodological issues specific to that research question. The five sample work plans in this report address the highest-priority research topics related to driver behavior (distraction, alcohol use, and fatigue) and infrastructure characteristics (inter- section crashes and interaction of roadway features and lane departures). Each sample work plan provides a high-level summary of a possible approach to the research question, including the rationale for examining that question, an outline for the data collection plan, and a descrip- tion of the analysis method. The sample plans do not provide an exhaustive description of the research approach, nor do they represent the ideal research approach. Instead, they outline some of the more important considerations and basic elements necessary to describe research using the SHRP 2 Safety project data. These plans also illustrate some of the fundamental chal- lenges that researchers must address when using naturalistic driving data to answer driving safety research questions. These sample plans may also be used by researchers as guides to the type of key informa- tion that may be required in S08 proposals. For example, the large amount of data collected as part of the SHRP 2 NDS will require careful attention in terms of how it is sampled, seg- mented, and aggregated. Equally important are the statistical methods used to extract policy- and design-relevant information from the data. Ideally, proposers to Project S08 will clearly Executive Summary
2explain their rationale regarding the selection of variables and data manipulation. Several other elements that should be included in the proposals, such as the analytical approach and potential pitfalls and limitations and how to address them, are explained in greater detail in this report. Finally, because crashes are rare occurrences, only a small number of crashes will be recorded in the NDS database. Thus, crash surrogates will be an important component in many analyses. The SHRP 2 NDS data will be useful in identifying crash surrogates that can be proven to be sensitive to the same contributing factors as actual crashes.
