One presents a statistic evaluation of Diniz metodology [2], to obtain uniformly probable response espectra, UPRS, based on the determination of the probability distribuition of the response extreme values of a single- degree of freedom structure, under a ground movement caused by an earthquake. Such movements are characterized by power spectrum density functions, each one of them representing a family of seismic events. On this way, one considers a power spectral density function, PSD, as a basis for the evaluation. Using the PSD, one generates artificial ground acceleration functions with the program SISMOSINV, developed within this work. From these artificial ground acceleration functions one obtains the response spectra with the SAP2000 program. Then, the UPRS are calculated, using the spectral value distributions, by adequate statistical techniques, implemented for the case. One then compares the UPRS obtained statiscally with those obtained by the probabilistic methodology and conclusions and recommendations are formulated concerning the applicability limits of this methodology One still compares the statiscally obtained UPRS with the design response spectrum prescribed by the USNRC [13] to evaluate the compromise between this spectrum and the target power spectral density function, TPSD, prescribed by the USNRC [1].