OBJECTIVE： To analyze influential factors of health utility evaluation， so as to provide reference for policy-makers， pharmaceutical companies and other relevant organizations prioritized in improving the quality of life for specific groups of people， and to provide supports to researchers to find the appropriate people to elicit the preferences for health states. METHODS： 240 urban respondents and 240 country respondents aged more than 16 were selected from Guizhou province. Utility values for a subset of health states defined by the EQ-5D descriptive system were directly elicited by applying Time trade-off technique. Linear regression models were used to estimate the effect of individual characteristics as age， gender， race， educational level， insurance status， health conditions on all health states， mild， medium and severe states. RESULTS： A total of 481 questionnaires were sent out and 467 valid questionnaires were collected with the effective rate of 97.09％.  Higher valuations were given by respondents who were the female， belonging to the minority group， elder person， with health insurance， with chronic disease diagnosed by the doctor， or with a poor health status at the past month. However， the respondents living themselves， with low educational level assigned lower valuations. CONCLUSIONS： Utilities for health states are influenced by respondent’s characteristics significantly. It is very important for the research to choose the appropriate people to elicit the preferences for health states.