摘要：We assess Canadian’s risk perceptions for genetically modified (GM) food and probe influences of socio-economic, demographic and other factors impinging on these perceptions. An internet-administered questionnaire with two stated choice split-sample experiments that approximate market choices of individual grocery shoppers is applied to elicit purchase behavior from 882 respondents across Canada. Data are collected to assess the influence on respondents’ choices for a specific food product (bread) of 1) product information which varies in content and by source and 2) information provided through labeling. These data also enable: a) analysis of trade-offs made by consumers between possible risks associated with GM ingredients and potential health or environment benefits in food and b) assessment of influences on respondents’ search for/access of product information. We rigorously document the extent and type of variation in Canadian consumers’ attitudes and risk perceptions for a selected GM food. This is pursued in analysis of experiment 1) data using a latent class model to analyze 445 consumers’ choices for bread products. We identify four distinct groups of Canadian consumers: 51% (value seekers) valued additional health or environmental benefits and were indifferent to GM content; traditional consumers (14 %) preferred their normally-purchased food; fringe consumers (4%) valued the health attribute and could defer consumption. Another 32 % (anti-GM) strongly opposed GM ingredients in food irrespective of introduced attributes. Thus there is a dichotomy in Canadian attitudes to GM content in food: a small majority of the sample (55 per cent) perceive little or no risk from GM food, but this is strongly opposed by 46% of respondents. Differences in gender, number of children in the household, education, and age are associated with the likelihood of segment membership. We also report on the search for information on characteristics of the GM food by a sample of 445 respondents with opportunity for voluntary access to related information through hyperlinks in the survey. Slightly less than half actually sought such information. Gender, employment status, rural or urban residency and the number of children in the household all affected the probability that respondents would access information. A further research component examines product choices made in the context of two common GM labeling policies: mandatory and voluntary labeling. We find these two types of strategies to have distinctive impacts on consumers and on measures of social welfare. Knowledge of these may help policy makers to make more informed analyses of the alternative labeling policies. Specific findings also provide base-line measures of Canadians’ attitudes to risks of GM technology in the context of food and environmental risks, as well as documenting the importance of context influences and reference points on consumers’ preferences for GM food. We also develop methodological improvements for accurately estimating the value of information on a negative attribute. The project built upon initial findings from a previous AARI project (#AARI Project #2000D037) and is complemented by research supported through a Genome Prairie GE3LS (Genetics, Ethics, Environment, Economics, Law and Society) project: “Commercialization and society: its policy and strategic implications.”
