Copula-based regression models have often been used for the analysis of clustered multivariate survival data. In particular, the Archimedean copula function is useful for modeling such dependency. For likelihood inference, one- and two-stage estimation methods are usually used. In this study, we focused on the behavior of the estimated model parameters against misspecification of Archimedean copular models. For this purpose, we investigated this sensitivity using simulation studies and real data analysis. The simulation results demonstrate that the estimated regression parameters are robust against misspecification of the copula model under the two-stage procedure but that they are not under the one-stage procedure.