The purpose of this study was to determine the cross-validation of a structural model among four new farming entry barriers and on-farm performances, and identify farm successors' entry barrier premium against startup farmers. For the purposes, survey data was gathered from 126 startup farmer and 185 farm successors, and the data was analysed by a structural equation model. The study classified new farmers into two groups(startup farmers/inherited farmer), and examined invariance of the model across the groups. In addition, the intercepts of endogenous latent variables including social network, professional competency, economic capital and market were compared. Based on the results from a multi-group analysis, first the model verified configural invariance and invariance of measurement weight, measurement intercept, and structural weight. Consequently, the cross-validation was supported. Second, social network affected the other entry barriers, and also had a relatively high indirect effect on on-farm performance. It implied that social capital formation was particularly important for setting up new farmers. Third, with social network held constant across groups, the entry barrier score of farm successor group increased professional competency score by 0.358 and economic capital score by 0.426, but decreased market score by 0.489. This result might be assumed for family farm propensity and farming background.