Uncertainty analysis is widely used in today's electromagnetic compatibility simulations in order to include the variation and tolerance caused by realistic nonidealities. However, previous research has neglected to investigate the validity evaluation of uncertainty analysis results. In this paper, a novel validity evaluation method, named mean equivalent area method (MEAM), is proposed to quantify the difference between the reference data and the simulation results, particularly for validity evaluation of the uncertainty analysis results in computational electromagnetics. Comparing with other existing works or confirmed conclusion in published references, the performance of MEAM is presented. Furthermore, in order to improve both efficiency and accuracy of the uncertainty analysis method (Monte-Carlo method or generalized polynomial chaos method), a scheme based on MEAM is presented to examine whether the uncertainty analysis method has reached convergence. Finally, the selection of the reference data and the limitation of MEAM are illustrated in detail in Section VI.