The seismic performance of buildings requires researchers to collect a suite of seismic records that are usually scaled to characterize the seismic hazard of the site. Unfortunately, most of the seismic signals in world earthquake databases originate from low-magnitude seismic events that must be scaled to reach a specific seismic intensity. After scaling the accelerograms, a nonlinear analysis of the buildings allows researchers to evaluate the expected performance and to assess damage limit states based on seismic demands in buildings' structural elements. Scaling seismic records is a frequent task in the nonlinear analysis of structures using different methodologies not always well justified in the studies. The scaling methodologies are based on several intensity measures, such as Peak Ground Acceleration (PGA), Effective Peak Acceleration (EPA), Effective Peak Velocity (EPV), and Spectral Acceleration at the building fundamental period, Sa(T1), among others. This study presents the effect of ten scaling methodologies on the expected behavior of reinforced concrete buildings subjected to a suite of accelerograms recorded in a high-seismic-hazard region. Based on the nonlinear analysis, the seismic demands of the buildings were assessed to determine the expected damage by selecting performance limit states. The results show an important variability in the building demands that can draw different conclusions as a function of the scaling methodology used in the nonlinear analysis.