This paper presents the results from an experimental project involving the full-scale testing of vehicles in a variety of maneuvers. The maneuvers include tripped rollover maneuvers resulting from impacts of eight test vehicles with both soil and curb pavement discontinuities. The reduced data from experimental tests form the impact condition variables which are augmented with vehicle design parameters in the analysis of results. The data are analyzed in datasets created according to such criteria as maneuver type, maneuver result, and type of terrain discontinuity. These results are investigated on the basis of mean values and coefficients of correlation.Results throughout the study support the conclusion that soil-tripped rollover maneuvers are qualitatively different than curb-tripped rollover maneuvers. The results obtained confirm that implementation of anti-lock braking systems and traction control systems reduces a risk of rollover when vehicle encounters a terrain discontinuity. Also covered is the processing of experimental results into a form suitable for computer simulation validation and the analysis of vehicle rollover behavior based upon obtained results.