Three new full-size partially encased composite shear walls were tested under cyclic loading by varying the thickness of the steel members (built-in steel plate, stiffeners, and steel tubes) and the number of stiffeners as variables. As the number of stiffeners increased, the horizontal displacement of the specimens exhibited an oblique layered distribution. Furthermore, increasing the thickness of the steel members led the horizontal displacement of the specimens along the height directions to change more linearly before yielding, but the specimen still showed bending failure. With the increase in the number of stiffeners, the displacement of the specimen along the height direction of the wall changed linearly throughout the loading process, and the stiffness degradation was mitigated. An increase in both the thickness of the steel members and the number of stiffeners increased the force transferred to the steel members. At the end of loading, the average strength degradation coefficients of the three specimens were 0.95, 0.95, and 0.92. The main specimen parameters were analyzed using OpenSees. The results showed that the axial compression ratio of the specimen should not exceed 0.3, and the thickness of the steel tubes should be controlled at 5 mm. In addition, the adverse effect on the seismic performance of the specimens was reduced when the shear-span ratio was greater than 2.43. Finally, the stiffeners and built-in steel plate had no significant effect on the seismic performance of the specimens when their thickness varied from 3 mm to 8 mm.