This research studied the problems and effectiveness of existing cross view school bus mirror systems. Interviews were conducted with 49 school bus drivers to ascertain their evaluations of the use and perceived effectiveness of various cross view mirror systems.Six commercially available cross view mirrors were randomly selected for testing. These mirrors were used to make the seven cross view mirror systems (each with 2 to 4 cross view mirrors) which were evaluated.The optical properties of each cross view mirror and the field of view (FOV) of each mirror system were measured in a laboratory environment. Mirror system FOVs were determined with the mirrors mounted on each of three different types of school buses. The directly observable FOV of each bus was also determined.Driver child detection field studies were conducted using eight bus drivers and six mirror systems with simulated children. Performance measures included child detection capability at selected locations around a bus as well as driver glance frequency and duration.The results of the field studies confirmed predictions from the laboratory FOV measurements and indicated wide differences in mirror system FOVs and subsequent driver performance.This paper is a condensed version of the reports written for this study. The reader should refer to (1)* and (2) for more details.