This study was to identify the determinants of inbound tourists' length of stay using the finite mixture negative binomial count model(FMM). The FMM model was also compared with zero truncated poisson(ZTP) and negative binomial model(ZTNB) Data used in this study were originally collected from 10,535 foreign tourists by Korea Tourism Organization in 2008. Results showed that the FMM was better than the others in terms of model fitness. In addition, results of the FMM model proposed two latent groups of length of stay, the 'short춖tay group' and the 'long춖tay group'. Furthermore, the determinants of length of stay in the two groups were significantly different from each other. Finding of study suggests the importance of doing similar kind of future researches for developing an effective inbound marketing plan of Korea.