Following the 2011 Christchurch, New Zealand earthquake, 60% of buildings in downtown Christchurch had to be demolished and reconstructed. Such practices are too costly both in terms of money and carbon emissions. Hence, structures need to be proportioned to survive more than one strong ground motion. To do so, it is important to understand the effect of previous shaking history on drift demands. Past studies conducted to investigate the effect of previous shaking on structural response were based almost exclusively on numerical analyses. To date, results from these numerical analyses have not been systematically vetted against measurements from structures tested in the laboratory or evidence from the field. To address this shortcoming, the effect of previous shaking history on drift demands was evaluated using 1) measurements from 220 experimental tests of structures subjected to repeated simulated base motions, and 2) evidence from 31 buildings surveyed after the 1985 and 2010 Chile earthquakes. Data from the experimental tests and field surveys showed that: 1) Repeats of medium or high-intensity ground motions strong enough to cause yielding in their first occurrence are unlikely to produce large relative increases in peak roof and story drift demands in structures with stable force-displacement relationships not susceptible to brittle failures. 2) Larger relative increases in peak drift demands are likely to occur if the first motion (in a sequence of motions containing a pair of repeated motions) is mild enough not to cause cracking and/or yielding, but the second instance of the same motion (repeat) is preceded by larger intensity motions causing cracking and/or yielding.