Much has been done in recent years to relate various roughness statistics to rider comfort in terms of a serviceability index (developed by Carey and Irick). Much less has been done on correct evaluation of a true profile in terms of summary statistics for field comparison of vehicle operating costs and user comfort. Important work in this area has been done by Gillespie and Sayers at the University of Michigan, and additional work has been done by Hudson, Williamson, and McKenzie at the University of Texas. Work has also been done in Brazil by Queiroz and others. At least two statistics have been offered for summarizing roughness information: (a) the average rectified velocity, by Gillespie and Sayers, and (b) the root-mean-square vertical acceleration, by Hudson et al. The purpose of this paper is to examine and compare these statistics with particular emphasis on their potential effects on rider comfort and their use as standard calibration statistics for response-type road roughness meters. A complete parameter study is reported comparing these two statistics. The results of the study will be helpful in selecting useful analytical techniques for routine applications.