This paper aims to apply a new methodology in bearing capacity analysis, often referred to as random finite element method (RFEM). This method considers the variability of soil parameters within the finite element method (FEM) by generating a Gaussian random field for the parameters within finite elements. The local average subdivision method (LAS) was used in this study to generate the Gaussian random field. However, soil parameters are not, generally, randomly distributed within neighboring soil elements; they tend to be correlated over a distance. Thus, the correlation length, the distance over the soil parameters are correlated to each other, was considered in this study. The Monte Carlo simulation was done for bearing capacity problem and some statistical and probabilistic methods were applied for analyzing the results to get the failure probability of footing on clay. This study would help to understand the effect of variability of soil parameter by using RFEM; so that the safety issues of geotechnical design can be determined in terms of probability of failure. Refereed/Peer-reviewed