This study presents a ground-motion selection and scaling methodology that preserves the basic seismological features of the scaled records with reduced scatter in the nonlinear structural response. The methodology modifies each strong-motion recording with known fundamental seismological parameters using the estimations of ground-motion prediction equations for a given target hazard level. It provides robust estimations on target building response through scaled ground motions and calculates the dispersion about this target. This alternative procedure is not only useful for record scaling and selection but, upon its further refinement, can also be advantageous for the probabilistic methods that assess the engineering demand parameters for a given target hazard level. Case studies that compare the performance of the proposed procedure with some other record selection and scaling methods suggest its usefulness for building performance assessment and loss models. Copyright (c) 2012 John Wiley & Sons, Ltd.