Alexander Pollatsek - US grants

9725228
Konold

The investigation will study a number of the projects that NSF is funding which students collect, share, and analyze real-world data in the course of conducting scientific investigations. IN these projects to date, the data gathering and sharing activities have gone more smoothly than the data ansalysis, in which many students and teachers have struggled. Such difficulties pose a serious barrier to fostering authentic mathematics and science learning and to enabling shared inquiry among students and scientists. This research will study the nature of the data analysis problems to provide a specific information that can help current projects better support data analysis skills, and a research base that can inform the development of future data analysis projects, materials, software, and teacher development efforts. The study will focus on issues such as data authenticity, tractability, and accessibility to students and teachers. Major products will include (1) analyses of individual projects' data sets, including and assessment of the extent to which interesting and interpretable findings can be reached and suggestions on how to make this data more accessible to students, and (2) a technical report that summarizes major findings, includes representative case studies that highlight both the problems and possibilities of students analyzing real world scientific data and makes recommendations about the design of data sharing and analysis projects, including ancillary materials and tools.

DESCRIPTION (provided by applicant): The long term objective of the proposed research is to understand better the causes of crashes among teen drivers right after they receive their solo license and to use that understanding to design and evaluate training programs that will reduce whatever skill deficits are revealed. During the first six months of solo driving, the crash rate for newly-licensed teen drivers decreases by a factor of five (McCartt, Shabanova, and Leaf, 2003). The only drivers at greater risk of crashing are those eighty years old and older. Perhaps not surprisingly, automobile crashes are the leading cause of death for newly-licensed drivers (Centers for Disease Control and Prevention, 2004). The problem is an ongoing national tragedy that has continued for far too long. Standard driver education programs, which typically involve 30 hours of classroom instruction and ten hours in the vehicle (four hours of observation and six hours behind the wheel), have, until recently, been the primary way teens learned to drive. Unfortunately, evaluations undertaken over the last forty years have shown little effect of such programs on crash rates (Mayhew and Simpson, 2002). Graduated driver licensing (GDL) programs are one response to the problem. The GDL programs clearly reduce crash rates among 16 year olds. However, newly-licensed drivers obeying the law-alone and during the day-are still over-involved in crashes, sometimes being as much as eight times more likely to be in a crash than are older drivers (Langone, 2006). There are many reasons that standard driver education programs and GDL programs may not reduce crashes as much as might be expected among drivers obeying the law during the first six months of restricted licensure. Perhaps most obvious is that these efforts do not directly remediate three behaviors that have been hypothesized to be the primary causes of crashes among novice drivers: failures to (a) anticipate hazards, (b) maintain attention, and (c) control speed appropriately. Unfortunately, little is known about whether these differences do indeed exist and, if so, the exact reasons newly-licensed and experienced drivers may differ from one another in each of these three areas. Until such is known, one cannot design training programs that remediate the deficiencies. We are proposing a four year program of theoretical and applied research to address these problems. Phase 1: In the first year, we will assess the hazard anticipation, attention maintenance and speed control skills of newly-licensed drivers (teens who have had their restricted license six months or less) on a driving simulator and compare their performance with experienced older (40-55) drivers. Eye movements, head movements, driver behaviors (e.g., foot on or off the accelerator, brake, etc.) and vehicle behaviors (e.g., velocity, brake pressure) will be monitored. We will determine the extent to which the differences in hazard anticipation, attention maintenance and speed control are due to differences in the knowledge, basic vehicle handling skills, and task management routines of the two groups of drivers. Phase 2: In the second year we will develop and evaluate on a driving simulator a PC-based training program designed to improve hazard anticipation, attention maintenance and speed control skills of newly-licensed drivers. The evaluations will take place immediately, one month, and three months after licensure. Phase 3: In the third year, we will repeat our evaluation of the PC-based training program, only this time evaluating newly-licensed drivers'hazard anticipation, attention maintenance and speed control skills on the open road immediately, one month and three months after training. Again, eye movements, head movements, and driver and vehicle behaviors will be monitored. PROJECT NARRATIVE: Automobile crashes are the leading cause of death for newly-licensed drivers (Centers for Disease Control and Prevention, 2004). During the first six months of solo driving, the crash rate for newly-licensed teen drivers decreases by a factor of five (McCartt, Shabanova, and Leaf, 2003). The long term objective of the proposed research is to understand better the causes of crashes among teen drivers right after they receive their solo license and to use that understanding to design and evaluate training programs that will reduce whatever skill deficits are revealed.