Lusk et al (2003) did a cross national survey in 2000 with 566 valid observations in the US [30]. According to the survey, US consumer had the least knowledge and least concern of GM safety compared with France, German or the UK back in 2000, and was the most sensitive to the change in price and the good consumer experience provided by GMOs.
Burton et al (2001) did an investigation of the consumer attitude towards GMOs in UK with 1626 valid observations [31]. To eliminate unwarranted prominence resulted by “a topic as contentious as GM food”, the investigators carefully avoided letting consumers guess the purpose of the investigation; instead, they hid it in the context of food system, and the attitudes to GM foods were actually inferred by carefully constructed models. The survey identified three groups of consumers: the infrequent, occasional, and committed consumers of organic food, which have different responses towards GM foods. The authors analyzed them with a heavily tested choice model which yielded many interesting results, such as that GM food was only one albeit very important concern of food safety, consumers calculate the willingness to pay to avoid the GMOs, and that the difference in attitudes differ significantly towards modified plants and introduced genes from other organisms, etc.
Bredahl did a cross-national survey of 2031 samples on the consumer attitudes and purchase intentions of GM yogurt and beer in Denmark, Germany, Italy, and the United Kingdom [32]. Estimated error (which passed the test) caused by language translation was taken into account. The result showed that the Danish and Germans had the most negative attitudes, the Italians the most positive, with the English in between. Very noteworthy is that the appendix of the essay was a list of elicited beliefs in the attitude model, including product and process related benefits and risks, as well as the normative, control, difficulty, and moral beliefs.
A face-to-face survey of 634 samples was conducted in Wuhan by Professors Liu Pengcheng et al (2004) of Huazhong Agricultural University in year 2004, to investigate the attitudes and expectations of consumers towards GMO [33]. From their report we can see, the sampling frame was carefully designed and cheap talk was employed to reduce bias.
Cheap talk is a special type of signaling game in the decision theory where the information is incomplete [34]. Incomplete means at least one player does not necessarily know the types, strategies or payoffs of the other players [35]. Cheap talk is the kind of signal given to the receiver by the information possessor that requires almost no cost. Cheap talk scripts which are lines that contain explanations of a suggestive nature that have been proven to be effective only to those who already agree with certain ideas suggested by the scrip. Cheap talk script is usually used in surveys to eliminate bias [36].
When investigators implied that GMOs can reduce pesticides, are more nutritious, the ratio of ``very willing'' to buy were 30.8 and 40.6; when investigators implied that GMOs might have unforeseen consequences to human health or have unforeseen influences to nature, the rate was reduced to 2.4 and 2.6. And about a quarter of consumers were impassive towards the investigation and were either ignorant of GMOs or didn’t care about the risks. According to their report, the consumers in Wuhan exhibited much ignorance of GMOs in year 2004, and the tendency to doubt increase with the increase of education background.
Carlsson et al (2007) did a mail survey of 1600 samples in Sweden in year 2004 [37]. They also used a cheap talk script to eliminate bias. The result mainly explored the marginal willingness to pay (WTP) between a labeling system and a ban on GM fodder, and it showed that Swedish consumer were willing to ``pay a significantly higher product price to ensure a total ban on the GM fodders'', and that their general attitude towards GMOs were pretty negative.
Chen et al (2007) did a consumer survey of 564 useful samples on consumer attitudes in Taiwan, China in 2003 [38]. The data was subject to heavy testing, and factors such as knowledge, trust, perceived benefits and risks were measured. Several results coincided with Liu Pengcheng’s survey: that the concept of GMO were relatively novel to people, that with the increase in knowledge come increased doubts, etc.
And there are still many, such as surveys done in Norway [39], China [40] or the developing world in general [41]; and comparisons such as the drastic difference between EU and US [42].
What I want to say is that these surveys all used various kinds of measure to eliminate what bias people might have about GMOs, and what they did was only a faithful account of their observations. The phrase market failure might have been mentioned somewhere in the analyses and it also exhibited the kind of bias I mentioned earlier (will look into later), but the bias must have been a subconscious one, since they really went to great lengths to eliminate bias and unwarranted tendencies that they realized exist.
"
