*4.1. Survey Administration*

The survey contained questions designed for the experiment and socio-demographic inquiries including gender, age, educational attainment, occupation, child, and income. The respondents were first provided with some detailed information on GM soybean imports and their connection to public interest. Each attribute was interpreted by the enumerators to make sure the respondents understood the survey. All the nonpaymen<sup>t</sup> attributes and price attributes in (Table 1) were shown to the respondents, who were asked whether they were willing to pay for the mandatory enhanced labelling of the GM soybean oil. They were also shown a sample of a CE scenario (see Figure 1) before the start of the experiment. We showed them what it would mean if "Product A" was chosen. The prices of all kinds of "X" brand 5-liter edible oil are shown in (Table S1), which were also presented to the respondents in the experiment. Two versions were developed to reflect the differences in the CE scenarios. Each respondent only took one version of the survey, assigned to six choice scenarios.

Adhering to the stratified random sampling, a self-administered questionnaire was utilized to collect data in the provincial capitals of the eastern, central and western regions. The eastern regions include Jinan of Shandong province, Nanjing of Jiangsu province, Shanghai, and Guangzhou of Guangdong province. The central regions include Changchun of Jilin province, Zhengzhou of Henan province, Hefei of Anhui province, Wuchang of Jiangxi province. The western regions include Lanzhou of Gansu province, and Guiyang of Guizhou province. This investigation was conducted at supermarkets and large-scale shopping malls by sixty-four undergraduates from Nanjing Agricultural University in 2017. Different social classes were sampled to avoid sample selection bias owing to sampling at a single site [48]. The survey enumerators approached potential respondents and invited them to participate if they wanted. The following steps were followed: (1) each respondent was confirmed to be an urban resident; (2) the selected respondents had food purchasing experience; (3) soybean oil was the family's main edible oil and was obtained through purchasing rather than through squeezing their own beans. After completing the survey, each respondent was offered a RMB 10 gift.
