The electrical resistivity of concrete is a key parameter for evaluating its durability, internal structure, and permeability. However, the conventional van der Pauw (vdP) method, primarily applicable to thin-sheet specimens, introduces significant size-effect errors in the resistivity testing of thick concrete structural members. Therefore, an integrated approach of laboratory experiments and finite element simulation were employed to systematically study the coupled effects of electrode arrangement, specimen dimensions, and electrode parameters. The observed size effects were parameterized using the identifiable geometric ratios L/w and l/d, leading to the development of a correction formula with defined applicability boundaries, where L is the cross-sectional elongation, w is the electrode width, l is the electrode length, and d is the sample thickness. The results show that the measurement error of resistivity caused by planar position deviation of electrodes is less than 0.75%, and the stability error of 28-hour continuous testing is less than 1%. There is a significant negative correlation between concrete resistivity and saturation, providing a quantitative indicator for permeability evaluation. When the cross-sectional side length L of the specimen is greater than 5 times the electrode width w, accurate measurement results can be obtained. If L≤5w, an equivalent thickness correction can be performed by introducing the fitted correction coefficient (k). According to the ratio of electrode length to specimen thickness, the current field is divided into three regions: point electrode approximation region, electric field expansion transition region, and size-independent stable region. A piecewise correction formula is proposed, and the average relative error between the corrected calculated values and measured values is 1.51%. The findings provide a new method for the resistivity testing of thick concrete.