OBJECTIVES: This study aimed to estimate Korean preference weights for EQ-5D-5L based on values elicited from representative sample applying EuroQoL standard protocol. METHODS: Of 1,085 general populations were recruited using multi-stage quota sampling method in Korea. Each respondent valued 10 health states using the composite time trade-off and 7 health states using discrete choice experiment in computer-assisted face-to-face interview. A range of prediction model was explored, the most appropriate model was determined in terms of goodness of fit, logical consistency and parsimony RESULTS: Model with dummy variables of the level of severity associated with each dimension, an intercept associated with any move away from full health, and a term that picked up whether any dimension in the state was at 4^th or 5^th severe level was best predicted the utilities for observed health states. The model was selected as the final model because all coefficients were statistically significant and logically consistent and it was parsimonious. The final model had a mean absolute error of 0.027 and none out of 86 exceeded 0.1 of absolute error. CONCLUSIONS: The final model in this paper appeared to predict the utilities of the states, which was valuated directly. This could be used interpolate quality weights of all EQ-5D-5L health states.