This paper assesses the forces driving farm exits over the period 1998 through 2002, using Korean farm-level data, focusing on basic farm and operator characteristics such as farm size and operator age likely to affect exits. This study applies a logistic regression model to the data set to investigate factors that contribute to farm exits and estimate exit probabilities for farms with different characteristics. The results show that exit probability initially declines with operator age until it reaches a retired age but increases after the age. The findings also indicate that the probability of farm exit decreased with very smaller farm size, but then increased for farms with medium sized, and again decreased for larger farm size. Farms with a relatively large number of family members are less likely to exit.