Purpose The EQ-5D-5L was developed to compensate for a high ceiling effect and lack of descriptive richness of the EQ-5D-3L. We evaluated psychometric properties of EQ-5D-5L in the general population. Methods Six hundred of adults were sampled from the general population in South Korea using a multistage stratified quota sampling method. Participants completed the EQ-5D-5L, EQ-5D-3L, and SF-36v2. One hundred participants were resurveyed for reliability evaluation. The ceiling effect, known-groups construct validity, convergent and discriminant validity, and reliability of EQ-5D-5L were evaluated. Results A smaller proportion of participants answered ‘no problem’ to all dimensions of EQ-5D-5L (61.2 %) than EQ-5D-3L (65.7 %, p < 0.01), indicating a reduced ceiling effect. Female, elderly, low-educated, and low-income participants reported health problems more frequently, indicating known-groups construct validity. The mobility dimension of EQ-5D-5L was better correlated with the physical component score (|r| = 0.48) than the mental component score (|r| = 0.25) of the SF-36v2, and the anxiety/depression dimension was better correlated with mental component score (|r| = 0.45) than physical component score (|r| = 0.34), indicating convergent and discriminant validity. The intraclass correlation coefficient of EQ-5D-5L index was 0.75. Conclusions The EQ-5D-5L has a smaller ceiling effect than the EQ-5D-3L and is a valid and reliable instrument to measure health-related quality of life in the general population.