A method is presented for computing a statically complete structural flexibility matrix from a dynamically measured flexibility matrix. When computed from a limited set of measured modal data, dynamically measured flexibility cannot reproduce the correct static force-displacement relations of the structure, i.e., it is not ''statically complete.'' A previously developed algorithm is used to include the effects of residual flexibility in the dynamically measured flexibility matrix, so that certain entries in the measured flexibility matrix can be considered to be statically complete. The method presented in this paper computes the remainder of the entries in the statically complete flexibility matrix by first forming a static flexibility matrix using assumed stiffness parameters and elemental connectivity, then scaling it such that it approximates the corresponding statically complete entries in the measured flexibility matrix. The method requires the solution of linear systems of equations only. The method is derived and applied to both numerical and experimental measured flexibility matrices, and the improved accuracy of the static flexibility over the dynamically measured flexibility is demonstrated.