*4.2. Descriptive Statistics*

This analysis is based on 804 samples collected in the survey. The samples in the eastern, central and western regions account for 45.02%, 35.07% and 19.90%, respectively. Specially, there are 2172, 1688, and 960 choice scenarios in the eastern, central and western regions, respectively, because each respondent responds to six choice scenarios. A statistic summary of socio-demographics of the sampled urban consumers is shown in (Table 2). Compared with the population, i.e., the national urban and rural residents, from the 2017 China statistical yearbook, the sample includes fewer males and shows a better education attainment, with 63.18% attending a professional school or holding a college or higher degree. The sample includes more young people under 45 years old. In addition, about 8% of the respondents have a job relating to biotechnology, and 56.47% of the families have minors. The average monthly household disposable income roughly follows a normal distribution, with the categories RMB 4001–6000, RMB 6001–8000, and RMB 8001–10,000 accounting for 31.84%, 16.92%, and 12.69%, respectively.


**Table 2.** Definition and descriptive statistics of the demographic variables.

a Standard deviation in the parentheses. b The population includes urban and rural residents based on information of 31 provinces from "2017 China Statistical Yearbook". b RMB 6.80 = USD 1.

> Table 3 shows a statistical summary of the variables used in the ML regressions. The consumers who are willing to pay for the enhanced mandatory labelling of the GM soybean oil account for 67.40%, 58.90%, and 57.50% in the eastern, central and western regions, respectively. From the nationwide perspective, the means of the three attributes are −0.087, −0.044, and −0.011, respectively. The average of the prices is RMB 35.21.



a N means the number of options. Standard errors in the parentheses.
