In recent decades, magnetic shields have provided basic experimental environments for the measurements of extremely weak magnetic fields represented by the biological magnetic signal. Excellent shielding performances, including the low residual field and high shielding factor, are necessary to ensure the quality of these weak magnetic signals and avoid the interference of external magnetic field. The magnetic shielding performance of the same device can be affected by the different degaussing and test conditions, which remains systematically studied. In this paper, the experiments with variable magnetization conditions, including different degaussing orders, test fields and environmental fields, are established in a nearly zero-field space to simulate the different situations during measurement. The residual field and shielding factor of the cubic shielding device are tested in these cases. Meanwhile, these shielding performances are analyzed from the perspective of the magnetization state and calculated based on the magnetic properties which are tested and fitted by Jiles-Atherton model. The results show the influence of these different conditions on the shielding performances of the cubic device, consistent with the numerical calculation. Under the same environmental field, the different degaussing order and test field lead to completely different residual field and shielding performance, respectively. The influence of the earth magnetic field on the shielding factor can be ignored due to its tiny equivalent bias field determined by the anhysteretic magnetization curve.