The size of spread footings was found to be unnecessarily large from some actual engineering practices constructed in Taiwan, due to the strict design provisions related to footing uplift. According to the earlier design code in Taiwan, the footing uplift involving separation of footing from subsoil was permitted to be only up to one-half of the foundation base area, as the applied moment reaches the value of plastic moment capacity of the column. The reason for this provision was that rocking of spread footings was not a favorable mechanism. However, recent research has indicated that rocking itself may not be detrimental to seismic performance and, in fact, may act as a form of seismic isolation mechanism. In order to clarify the effects of the relative strength between column and foundation on the rocking behavior of a column, six circular reinforced concrete (RC) columns were designed and constructed and a series of rocking experiments were performed. During the tests, columns rested on a rubber pad to allow rocking to take place. Experimental variables included the dimensions of the footings, the strength and ductility capacity of the columns and the intensity of the applied earthquake. Experimental data for the six circular RC columns subjected to quasi-static and pseudo-dynamic loading are presented. Results of each cyclic loading test are compared against the benchmark test with fixed-base conditions. By comparing the experimental responses of the specimens with different design details, a key parameter of rocking behavior related to footing size and column strength is identified. For a properly designed column with the parameter higher than 1, the beneficial effects of rocking in reducing ductility and the strength demand of columns is verified.