ABSTRACT Uncertainty in neutron dose after penetrating a 3-m-thick concrete was estimated based on three different kinds of Total Monte Carlo methods under random sampling methodology. A thousand random nuclear data files were generated for 28Si by T6 by perturbing underlying model parameters. In the first method, these files were used directly to yield processed library preserving all the correlations among different physical quantities. In the second method, a covariance file in the ENDF-6 format was generated and 1000 random files were produced based on the covariance file. In the third method, the random files populated by T6 were used but the angular distribution data were kept fixed to non-perturbed nominal ones. It was found that the second and third methods gave equivalent variance of the neutron dose after deep penetration of a concrete, and this uncertainty was larger than the uncertainty given by the first method. It shows that the correlation between total cross section and angular distribution of elastic scattering affects uncertainty of the calculated neutron dose. It could be concluded that the uncertainty obtained by using the covariance files given in the ENDF-6 format may not give correct results for the uncertainty of deep penetration calculation.