A direct displacement based design (DDBD) methodology for concentrically braced frames (CBFs) is validated using a combination of experimental and numerical data taking into account the behaviour of brace-beam-column connections. Twelve full-scale shake table tests with four different brace cross-section sizes, two different gusset plate design methodologies and two different brace connection configurations are used to evaluate the design procedure at three different performance levels (continued operation, life safety and collapse/near collapse) through successive scaling of test excitations.It is found that the DDBD procedure can predict the behaviour of the shake table CBFs for the three levels of excitation. Accurate results have been found for life safety design where the mean ratio and coefficient of variation of base shear (F-b) calculated using the DDBD method to the values found from shake table tests are 0.96 and 0.17, respectively. Furthermore, the mean ratio and coefficient of variation of brace area (A(b)) calculated from the DDBD method to the values used in shake table tests are 1 and 0.08. respectively.The design procedure was also verified using nonlinear time history analysis (NLTHA) using a selection of earthquake records with displacement spectra compatible with those of the exact accelerograms recorded during the shake table tests. At the life safety performance level, the mean peak displacement demand and base shear values obtained from NLTHA were observed to be 82% and 91%, respectively, of the design values determined by the DDBD methodology. (C) 2018 Elsevier Ltd. All rights reserved.