CONTENTS

• Introduction
• Consumer Focus Groups
• Product Attribute Manipulation Using Conjoint Analysis
• National Survey
• Part One - attitutes to genetic engineering in food production
• Part Two - Intention to purchase or avoid tomatoes genetically modified to reduce the need for pesticides.
• Growers and Handlers

INTRODUCTION

This report outlines preliminary findings from a three year research programme, funded by the Foundation for Research, Science and Technology, investigating the New Zealand public's perceptions of transgenic plant products. The research was carried out by HortResearch, in collaboration with AgResearch, Forest Research Institute, Auckland University and Massey University.

Stage One of the research involved consultation with 41 different stakeholder groups to ensure that the information collected would be relevant and useful for policy and planning decisions. The Delphi Consultation process was used to pool insights and ideas.

Despite the disparity of the 41 organisations and industries interviewed, consensus was gained very quickly. Respondents requested information covering a wide range of issues, indicating a need for some preliminary research prior to the initiation of any quantitative study.

Focus groups provided an indepth understanding of the issues among consumers, growers and handlers. A product attribute manipulation study assessed the influence of genetic engineering on consumers' purchase intention. Finally, a national survey was used to measure and explain the attitudes of a representative sample of New Zealanders.

These results are preliminary findings only. Further analysis is required. The survey in particular was designed to allow good correlation between different series of questions to give a more in-depth understanding of the respondents. The survey has also been conducted in European Union countries and this will allow comparisons to be made with responses in these countries.

CONSUMER FOCUS GROUPS

A total of 6 focus groups were held with three types of respondents: `general consumers' (n=15), those with affiliation to environmental groups (n=11) and Buddhists (n=20). However, there was subsequently found to be very little difference in issues raised between the various group types.

The following are key issues identified in these discussion sessions:

• Genetic modification is strongly associated with perceived 'unsafe' treatments of food such as pesticides and additives. This is of particular concern in the context of food such as fruit and vegetables that are bought for their nutritional contribution to the diet.
• Desire for labelling depends on product category, occasion and consumer. For example, consumers purchasing a `healthy' item for their family were more likely to be interested in labels than if they were purchasing a chocolate bar as a snack for themselves.
• There are concerns about the unknown short and long term risks on health and the environment.
• There is a perceived lack of choice and control over consumption of genetically modified food, due to the lack of labelling regulations, and the resulting perception of being 'part of an experiment' without having given consent.
• Big businesses are perceived to have a monopoly over the distribution of information, and policy/regulation formation. Hence there is a perceived lack of regulations and objective information available to the consumers.
• Awareness of the issue of genetic engineering is very high, but most people admit they understand very little about it. To alleviate a sense of `information overload', information has to be very simple and easily accessible. Leaflets could be successful, if distributed by an organisation (such as ANZFA) that people recognise as a 'watchdog' group, especially if accompanied by an 0800 number.
• Because consumers are reliant on particular groups (ANZFA, governmental) to provide them with the required information, a high degree of trust in those groups is required for the information to be believed.

Focus groups are an excellent tool for exploring underlying issues involved in a poorly understood concept. However, they cannot provide an indication of the prevalence of opinion among a population, nor the strength with which it is held. Quantitative research thus compliments this approach. The quantitative research was only conducted with consumers.

PRODUCT ATTRIBUTE MANIPULATION USING CONJOINT ANALYSIS

Conjoint analysis is a research method that systematically manipulates the attributes of a product to determine the relative importance of those attributes in the overall purchase decision. It also allows for identification of various segments of consumers in the market. Three attributes (price, benefit and technology) were evaluated on two product categories (health food (tomatoes) and snack food (chocolate biscuits)) by 115 respondents in Auckland.

Although there were similiarities in the patterns of behaviour among those presented with tomatoes or chocolate biscuits, there were also some interesting differences between the two.

Many consumers used similar decision-making strategies when presented with tomatoes or chocolate biscuits.

These people fell into four clusters:

• Cluster one (7-13% of the sample) were very price sensitive, preferring a cheap genetically modified product over an expensive organic product. In addition, the members of this group tended to be relatively happy to try new products, had lower environmental concern, and were less likely to see negatives associated with genetic engineering. Few of these respondents objected to, or worried about gene technology.
• The second cluster (25%) were also price sensitive, but not sufficiently so to compensate for the way the product was produced. As with the above group, members of this cluster saw relatively few negatives associated with genetic engineering and were less likely to say they worried about, or objected to the technology.
• The third cluster (21-30%) was strongly anti-genetic modification, preferring organic or all-natural products. Neither price nor benefit could compensate for the production method. They were less likely to believe the technology was beneficial and many of these consumers personally worried about, and objected to the technology.
• The fourth cluster (11-12%) was similar, but preferred specific guarantees that the product was not genetically modified. Members of this group tended to be relatively neophobic (averse to new products or situations). This could well reflect the desire by those members for reassurance, and a willingness to pay more for this.

For some respondents however, decision-making strategy differed between the two product categories. Interestingly, there were two groups of respondents for whom benefit made more of a difference to the rankings than did price, although not enough to overcome the influence of technology. Members of these clusters (11% and 7%) had a preference for pesticide-free products. Both clusters contained members who perceived genetic engineering relatively negatively, and were more likely to be male. In addition, some members provided relatively high environmental concern scores. Not surprisingly, these clusters only appeared when presented with tomatoes.

Benefits were not important for any of the clusters identified when presented with chocolate biscuits. This is perhaps not surprising, since the benefits offered by the biscuits were related to health - something that is not such an issue with snack foods.

The presence of a group for whom price does play a minor role suggests that increasing the price differential (it was comparatively small in this study) may have resulted in a larger price-sensitive group, and a smaller moderately price-sensitive group.

NATIONAL SURVEY

While product attribute manipulation can serve as a predictive tool in terms of the willingness of consumers to accept or reject genetic modification, it makes no attempt to measure or explain the consumer's attitudes. A survey was developed as part of a European Union project to do this. Joanna Gamble was involved in pilot testing this survey in the UK and adapted the survey slightly for use in New Zealand.

1600 surveys were sent out nationally as part of an omnibus survey distributed by CM Research. The nature of omnibus surveys allows a number of investors to place their own question sections into the survey, and share the costs of the project. Respondents were contacted by random digit dialling, and, to ensure a good cross section of people were interviewed, the person who last had their birthday was recruited. Respondents were quota-sampled on the basis of age, gender and region. 908 completed surveys were returned¹.

Part One - attitudes to genetic engineering in food production

A total of 64 questions were asked, and respondents were required to rate their opinion on a scale of 1 to 7.

• Over half the respondents indicated some negativity to genetic engineering in food production (giving a rating of 1-3), with a quarter feeling extremely negative (giving a rating of 1). Women were more likely to feel this way than men, as were people aged between 35-44 and over 65. In comparison, depending on the question wording, only 10-18% felt positive (gave a rating of 5-7) to any degree about the technology (men and those aged between 18-24). A third were undecided (giving a rating of 4).
• The extent to which a respondent agreed that there were benefits was the main determinant of whether the respondent would accept the technology in food production. Such benefits included²:
• Increased product choice (13% agree, 27% disagree)
• Environmental solutions (11% agree, 32% disagree)
• Reduced price (11% agree, 32% disagree)
• Improving standard of living of self/family (9% who agree, and 36% who disagree) and future generations (4% who agree, and 47% who disagree)
• Better quality (5% agree, 45% disagree)
• Improved healthiness (4% agree, 47% disagree)
• Typically, a far greater proportion of respondents tended to disagree that there were benefits than agree. This suggests that either the respondents did not believe that the technology could offer benefits, or they were not yet aware of the benefits.
• Risks assessed included:
• Unknown long term consequences on environment and human health (63% who agree, and 5% who disagree)
• Unnatural (52% who agree, and 10% who disagree).
• Interference with wild species in nature (38% who agree, 11% who disagree)
• Environmental hazards (30% who agree, 14% who disagree)
• Threat to human health (29% who agree, 18% who disagree)
• Allergies (19% who agree, 17% who disagree)
• The fact that the respondent agreed that there were risks (for instance, that the technology was a threat to human health) did not make the respondent less accepting of the technology. Rather, it affected the likelihood of whether they agreed there were benefits of the technology (for instance, improving health).
• Those who thought they personally knew little about the technology (37% gave a rating of 1-3 on both questions), those who thought authority (science, government and industry) was knowledgeable (5% gave a rating of 5-7), and those who scored relatively low on the attitude to nature scale (2%) were more inclined to believe genetic engineering offered a number of benefits, and less inclined to believe in the risks.
• Those who perceived themselves to be knowledgeable (2%), those who thought authority was unknowledgeable (27%), and those who had a high degree of environmental concern (53%), were more likely to perceive risks associated with genetic engineering, and less inclined to believe in the benefits.
• Actual knowledge of food technology issues was reasonably high, with more than half of the respondents scoring over 75% correct (6-8 correct out of 8 questions). In comparison, 9% scored less than 50% correct (1-3 correct). Those who really were knowledgeable were slightly more likely to see benefits of the technology, but the influence was less than that of perceived knowledge.
• Perceived and actual knowledge were relatively poorly correlated.

Part Two - Intention to purchase or avoid tomatoes genetically modified to reduce the need for pesticides.

A further 34 questions were asked about tomatoes genetically modified to reduce the need for pesticides. Once again, respondents were asked to rate their opinion on a scale from 1 to 7.

• About 40% of respondents said they would probably avoid the tomatoes (giving a rating of 1-3), 25% of whom would definitely avoid them (a rating of 1). Between 20-25% said they probably would buy the tomatoes (giving a rating of 5-7). A further third said they were undecided (a rating of 4). No gender differences were apparent, but rural respondents were more likely to indicate they would avoid the product.
• In considering whether to purchase or avoid tomatoes genetically engineered to reduce the need for pesticides, people were strongly influenced both by their own moral judgement of that behaviour, and by how they thought society would judge their behaviour.
• Urban dwellers gave higher social norm ratings (ie, they predicted their peers would be more accepting of the technology). Rural people were more inclined than urban people to believe that they could purchase or avoid genetically modified tomatoes if they chose.
• Respondents were asked how strongly they agreed or disagreed with a number of outcome statements. The biggest gaps that occurred between agreement and disagreement were in terms of the unknown long term consequences of eating genetically modified tomatoes (63% agreed, 5% disagreed), the lack of distinguishability from other tomatoes at the supermarket (46% agreed they would be indistinguishable, 6% disagreed), and the unfamiliarity of the product (41% agreed they were unfamiliar and 7% disagreed). Other questions included:
• These tomatoes are unnatural (12% disagreed, 38% agreed)
• These tomatoes are wholesome (26% disagreed, 18% agreed)
• I have no confidence in these tomatoes (17% disagreed, 31% agreed)
• I expect these tomatoes to have a good quality and taste (12% disagreed, 33% agreed)
• The genetic modification of this tomato benefits nature (34% disagreed, 9% agreed)
• I expect these tomatoes to have a nice texture (9% disagreed, 27% agreed)
• I expect these tomatoes to be easy to handle and use (8% disagreed, 30% agreed)
• My family thinks I should (29% avoid, 6% buy)
• I expect these tomatoes would be too expensive for me (11% disagreed, 25% agreed)
• These tomatoes would never be available where I do my shopping (34% disagreed, 14% agreed)
• I'm sure I would notice if these tomatoes were produced by the means of gene technology (33% disagreed, 16% agreed)
• The genetic modification of these tomatoes is morally wrong (14% disagreed, 31% agreed)
• The genetic modification of these tomatoes goes against my basic principles (15% disagreed, 34% agreed)
• The genetic modification of these tomatoes interferes with nature (11% disagreed, 36% agreed)

The survey was specifically designed to allow a high level of correlation between answers to different questions. Considerable testing of this model has been undertaken with good results. We will be conducting ongoing analysis of these correlations over the next few months and will make these results available to you. In addition, we plan to make comparisons with those found in Europe.

The initial findings support the suggestion by those involved in the 1996 Eurobarometer survey that acceptance of the technology would be strongly determined by moral considerations. The findings also support the prediction that genetically modified products with tangible benefits will be more acceptable than the current offerings. Finally, we have shown that consumers do have a reasonable understanding of the underlying science of the technology, and should therefore understand information that might be made available to them.

GROWERS AND HANDLERS

Focus groups were also conducted with potential growers and handlers of transgenic plant products to see whether the issues that were raised differed from those of consumers. These results will be included in the full report.

¹ Typically, around 6000 New Zealanders are contacted for this omnibus survey, resulting in a response rate of about 15%. A particularly low response rate was achieved for Maori, and for respondents aged under 34. A comparatively high response rate was achieved for those aged over 55. Maori respondents (n=36) tended to be less positive about using genetic engineering in food production (although this was not significant). Respondents aged 18-24 were the most positively inclined of all respondents towards the use of the technology in food production, while those aged over 65 were some of the least. Other methodologies may have resulted in higher response rates, and consequently better representativeness of New Zealanders, but financial constraints required us to use the current method.

² percentage agree are the percentage who rated a 6 or a 7, and the percentage who disagree are the percentage who rated a 1 or a 2.