The present research focused on evaluating an effect of training program for elderly drivers based on coaching method.　A before-after comparison of their driving performance, self-evaluation and instructors’ evaluation was conducted in three driving schools in Japan with 120 participants(117 male and 3 female), who were divided into a training group and a control group.　The driving performance, the driver’s face and traffic scene were recorded by in-vehicle data recorder and video cameras.　After driving through a driving school course, the participants in the training groups, usually four participants in each session, watched videotaped scenes of their own driving mainly at intersections and discussed their driving style.　Driving instructors supported the discussion with relevant comments and questions to the participants.　The training demonstrated a marked effect immediately after and a certain effect one month after the training on both driving performance and instructors’evaluation.　The participants demonstrated more searching head movements after the training than before the training.　They drove at lower speeds at uncontrolled intersections both with restricted visibility and with stop sign after the training.　The effects on driving performance were maintained even after one month.　Instructors’evaluations were improved immediately after the training and were stable even after one month.　Self-rated driving performance declined immediately after the training, but had almost recovered after one month.

In this study the behavioral characteristics of adult pedestrians when crossing a crosswalk were observed to identify factors linking age and victims of pedestrian traffic accidents, as shown in a previous study.　From August to October 2009, a total of 303 adults were observed using two separate pedestrian crossings in Oita and Beppu cities respectively.　The crossing in Oita was controlled by a pedestrian light, and the one in Beppu was not.　Firstly, the data was analyzed using Fisher’s exact test (extended).　The results showed a significant difference according to age, in ability to cross within the allocated time and adaptive speed control.　Secondly, multiple correspondence analysis method was used to plot the data on x and y axes.　This showed age to be a significant factor.　The criteria for the axes was then developed as follows: x＝attention and inattention ; y＝declining physical condition and recklessness.　Both characteristics of inattention and decline in physical condition were most populated by the group of 65 years and older.　The results suggest that inattention caused by mental and physical decline in senior pedestrians are causal factors in them being the highest group of victims of pedestrian traffic accidents.