Spectral regression studies of four selected research groups, namely, Boore et al. at the U.S. Geological Survey (U.S.G.S.) in Menlo Park, California, Abrahamson and Silva at the Pacific Gas and Electric Co. (PG&E) in California, Trifunac et al. at the Strong Motion Group of the University of Southern California (U.S.C.) in Los Angeles, and Ambraseys et al. at the Imperial College of Science, Technology and Medicine in London, England. Their scaling procedures are described, and their approaches are compared. All regression equations reviewed depend upon magnitude, distance, and local site conditions, though different groups used different database, definitions of magnitude, distance and site conditions. Among the many differences, one stands out: Boore and co-workers in 1997, Abrahamson and Silva in 1997, and later Ambraseys and co-workers in 2005 considered using fault-type scaling variables to differentiate among the motions from different fault mechanisms, while the USC Strong-Motion Group introduced a source-to-station path-type term.