This paper describes an approach to the development of a test procedure for measuring the combined steering and braking performance of automobiles with possible application to the formulation of a safety standard. The study was performed under Contract No. DOT-HS-4-00971 for the NHTSA. The influences and interactions of the many factors which must be considered in developing a meaningful and objective procedure are discussed. These include such diverse items as definition of test conditions and equipment requirements, identification of the significant operational variables (initial speed, permissible control functions, performance metrics, etc.), means for discriminating between vehicle and driver effects, and determination of acceptable performance levels. A full-scale test program, aimed at obtaining actual performance information on the effect of various factors, is described. Emphasis is placed on identifying operational situations having distinct for accident potential (i.e., conditions in which excessive brake pedal force might be applied as in a panic stop) and on the role of the driver in attempting to minimize loss-of-control possibilities at this condition. Test results showing the marked differences in directional behavior which can occur as functions of magnitude of applied brake pedal force (and, consequently, of wheel lock-up patterns) are presented in the form of graphs and tables. Comparisons between driver-controlled and machine-controlled operations have been made in order to devise a low-cost unambiguous procedure that properly reflects accident potential.