5. Approval of genetically engineered products

5.1 Measuring Options on Complex Issues
5.2 Introduction to Genetic Engineering
5.3 Levels of Approval
5.4 Structure of Opinion on Genetic Engineering
5.5 Summary: Approval 

5.1 Measuring Opinions on Complex Issues

In survey research there are no hard and fast rules about the best way to measure difficult concepts -- doing so is an art in itself. But the standard procedure, and the one that usually works best, is to ask a number of relatively specific, concrete questions and then average the answers. For example in a study of voting behaviour, to discover what voters think about government ownership, best practice is to ask a number of specific questions about ownership of particular industries (electricity generation, railways, steel manufacturing, cars, shops, farms, etc.) and then construct a combined 'government ownership' scale from the answers. Analogously, in a study of psychiatric depression, standard practice is to ask about a long list of specific symptoms and then construct a summary scale from the answers (Headey, Kelley and Wearing, 1993). Asking people a single, direct question is rarely the optimal approach, save occasionally in areas where people have firm, long-standing views on a simple black-and-white topic (for example, whether they support the Labor Party or the Coalition). Genetic engineering is not such a topic.

An example may make the logic of the standard 'multiple item indicator' approach clearer. Suppose you want to know how much people like French Post-Impressionist paintings. The ways you can ask the question depend on the knowledge and sophistication of the respondents

  1. If you were dealing with people who know a lot about art -- for example, people with a BA degree in art history -- a single, general question would suffice: "How much do you like French Post-Impressionists?" To answer this, respondents need to recall just who the Post-Impressionist painters were, recall what their paintings look like, and decide how much they like the paintings.
  2. But most non-specialist university graduates have at best only a vague idea about who the Post-Impressionists were, so it would be better to ask them a number of more specific questions naming the particular painters you are asking about: "How much do you like Bonnard's paintings?" "How about Derain?" Etc. Then one can get a good idea of what respondents think of Post-Impressionists by averaging their answers to the several questions. To answer these questions, respondents need to recall what Bonnard and Derain's paintings look like, and decide how much they like them.
  3. But even these questions assume a fair level of knowledge, specifically that respondents are familiar with (for example) a reasonable selection of Bonnard's paintings and so can say whether they generally like them or not; that they are familiar with a selection of Derain's paintings, and so can rate them; and so forth. This may (or may not!) be reasonable for the general run of university graduates. But for a normal, population with only year 9 or year 12 schooling, it is not reasonable. For most people, it would be much better to show several of Bonnard's paintings and ask how much they liked each of them; then show some Derains, and ask about them, and so forth. This does not require respondents to know which artists are Post-Impressionists, nor to recall any of their pictures. Instead it only requires them to look at pictures and decide how much they like them -- a far easier task.
We have therefore adopted the third style -- the conventional multiple indicator research -- in asking about genetic engineering. To answer these questions, respondents do not have to know which scientific developments they have heard about over the past few years involve genetic engineering, nor do they have to recall unaided particular genetic engineering projects that have been in the news over the past few years. They have only to read the questions and respond to the particular, concrete proposals in them -- a far easier task. In addition to clarity, asking questions in this third style has several important analytic advantages: We have therefore adopted the usual -- and generally correct -- multiple indicator strategy of asking about a number of particular, concrete instances of genetic engineering. The particular instances are among those already well into development in Australia and overseas (Australian Science and Technology Council, 1993).

5.2 Introduction to Genetic Engineering

The genetic engineering module required a lengthy introduction, because our pre-test suggested that many people had not heard of it, and for many others, a bit of reminding of what genetic engineering is about is useful to focus their thoughts, to remind them of things they may have already heard about (as most had), and to explain the topic to those previously unacquainted with it. The introduction:

Genetic engineering is a new way to create new products. Scientists can use genetic engineering on plants or animals to change things like their size, colour, or taste. They do this by moving a gene from one kind of animal or plant to another, or by turning a gene off.  

Our double goal with this introduction was to maintain scientific accuracy but also to make it readily accessible to respondents with little education. We began with this general sketch of the technique, and then proceeded to a particular concrete example, because people can always think better with a concrete example to hand:

 

Recently, scientists have made an improved variety of tomato that has a better texture, costs less, and might make a valuable export. They turned off one of its genes, which would otherwise have made the tomato go mushy. 

After introducing the example, the introduction raises the key theme of safety and danger that will echo through the rest of the examples in the module:

After careful study, a government regulatory committee believes that the new tomatoes are safe. Most scientists agree. But a few are worried and some nation-wide environmental groups say the tomatoes might be dangerous and should be banned. 

This scenario sets up the actual situation in the recent U.S. introduction of genetically engineered tomatoes, a scenario likely to be repeated in Australia in coming years. At this stage the question is hypothetical -- the tomatoes are not yet on the Australian market and Australian regulatory bodies have not been asked to evaluate them -- but hypothetical questions are perfectly normal in politics and elsewhere, and usually pose respondents no particular problems -- citizens are quite accustomed to deciding about policies that do not exist and may never exist (for example, the GST).

We closed the introduction with a reassuring sentence to re-affirm that we were seeking everyone's opinion, not just the opinions of people who consider themselves experts: "

Most people have not heard much about genetic engineering. We just want your opinion, your best guess.  

5.3 Levels of Approval

After introducing the topic of genetic engineering in this way, we then asked several questions (described later) about genetically engineered tomatoes, and concluded with a summary evaluation:

  

c. If clearly labeled, are these new tomatoes a good idea or a bad idea
                Very good idea                  17
                Good idea                       47              
                Mixed feelings, hard to say     28              
                Bad idea                         5              
                Very bad idea                    3                                                      
                                                ---
                                               100% (mean= 63)

We went on to ask about 7 other genetic engineering projects that are underway in Australia or overseas, or likely to be underway in the relatively near future. We began with a general introductory phrase "Here are some other things that scientists might make with genetic engineering...", and then asked people to rate the desirability of a list of possible genetically engineered products:  

Here are some other things scientists might make with genetic 
engineering...

e. A treatment that would save the lives of people who have blood cancer 

                Very good idea                  65
                Good idea                       29              
                Mixed feelings, hard to say      5              
                Bad idea                         1              
                Very bad idea                    1                                                      
                                                 ---                                                    
                                                 100% (mean= 89)

f. A genetically engineered drug that lowers blood pressure better 
than other drugs, reducing the risk of heart attack

                        Very good idea  ...etc... 
                                (mean 87) 


g. Genetically engineered cotton that resists insect pests -- this could 
greatly reduce the use of chemical pesticides

                        Very good idea  ...etc... (frequencies) 
                                (mean 87 ) 

h. Genetically modified viruses to protect farm crops by attacking insect 
pests, such as beetles and locusts

                        Very good idea  ...etc...  
                                (mean 76 ) 

i. Modified viruses to control imported animal pests (such as rabbits or 
feral pigs) by preventing them from breeding

                        Very good idea  ...etc...  
                                (mean 76 ) 

j. Leaner, healthier pork (assuming it is clearly labeled, so you can decide 
for yourself whether or not to buy it) 

                        Very good idea  ...etc...  
                                (mean 73 )
k. Healthier cooking oil and margarine, with more of the desirable unsaturated 
fats and fewer of the undesirable fats 

                        Very good idea  ...etc... 
                                (mean 79 )
(Questionnaire)
(Frequencies)

Very few people were unable to form an opinion: only 3% of the sample declined to answer these questions, on the average. That is about average for the questionnaire -- perhaps rather lower than average -- and well below the levels of "missing data" that occur on obscure topics (for example, in another ISSS, we asked people to rate their feelings towards the Chinese leader, Mr. Deng, and 25% declined to answer the question: Kelley, 1995b). In a paper-and-pencil format, there is no social pressure to answer any question one would prefer to skip, so the low level of missing data offers another sign that real public opinion on the topic exists.

5.3.1 Genetically engineered medicines

The public overwhelmingly supports trying to use genetic engineering to make "A treatment that would save the lives of people who have blood cancer." We offered five answer categories, and 64% of respondents declared that trying to use genetic engineering to make a treatment for blood cancer was a "Very good idea", and another 29% said a "Good idea". That makes 93% in favour. 5% had mixed feelings on the subject, 1% thought it "A bad idea" and another 1% thought it a "Very bad idea". Scoring these answer categories from 0 (a very bad idea) to 100 (a very good idea) gives us a summary "desirability rating". Australians, on average, rate trying to use genetic engineering to make a treatment for blood cancer at 89 points out of 100.

There is also overwhelming support for trying to make "A genetically engineered drug that lowers blood pressure better than other drugs, reducing the risk of heart attack". 59% of the public found this a "Very good idea", and 33% said a "Good idea". That makes 93% in favour. 6% had mixed feelings. 1% said a "Bad idea" and another 1% said a "Very bad idea". On average, Australians rate a genetically-engineered blood-pressure drug at 87 points out of 100.

5.3.2 Genetically Engineered Food

Using genetic engineering to make new foods also attracted substantial support, provided that the new foods would be clearly labeled.

The public, on average, rated at 79 points out of 100 "Healthier cooking oil and margarine, with more of the desirable unsaturated fats and fewer of the undesirable fats." This is a high rating, although not so outstandingly high as for the genetically engineered medicines. 40% said it was a "Very good idea", another 42% said a "Good idea", 13% have mixed feelings, 2% think its a bad idea, and 2% think its a "Very bad idea".

With a desirability rating of 73, "Leaner, healthier pork (assuming that it is clearly labeled, so you can decide for yourself whether or not to buy it)" is about three-quarters of the way towards being thought a "very good idea". 29% think it is a "very good idea", and another 42% think it is a "good idea". 19% have mixed feelings. 4% think it is a "Bad idea" and 3% say a "Very bad idea".

The genetically engineered tomatoes that served as the example in the introduction elicit a desirability rating of 67, still strongly favorable but noticeably lower than other foods. "If clearly labeled, are these new tomatoes a good idea or a bad idea?" 17% said a "Very good idea", 47% said a "Good idea". 28% had mixed feelings. 6% thought these tomatoes a "Bad idea" and 3% thought them a "Very bad idea".

5.3.3 Genetically Engineered Agricultural Products

"Genetically engineered cotton that resists insect pests -- this could greatly reduce the use of chemical pesticides" attract a desirability rating of 87 points out of 100. That is overwhelming support. Fully 59% of the populace endorsed them as a "Very good idea" and another 34% found them a "Good idea". 6% reported mixed feelings on genetically engineered cotton. 1% thought it a "Bad idea" and another 1% thought it a "Very bad idea".

The citizenry accords "Genetically modified viruses to protect farm crops by attacking insect pests, such as beetles and locusts" a desirability rating of 76 points out of 100. This is substantial support, although clearly lower than for the pest-resistant cotton. 41% thought these viruses a "Very good idea", and another 33% thought them a "Good idea". 18% had mixed feelings. 5% thought them a "Bad idea" and 3% thought them a "Very bad idea".

Australians have very similar views about "Modified viruses to control imported animal pests (such as rabbits or feral pigs) by preventing them from breeding": they give these contraceptive viruses, too, a desirability rating of 76 points out of 100. 40% say these contraceptive viruses are a "Very good idea", 34% think them a "Good idea". 17% had mixed feelings. 5% said these contraceptive viruses were a "Bad idea", and another 3% said a "Very bad idea".

5.3.4 Summary: Desirability Ratings

Thus, the citizenry holds positive attitudes towards this entire array of potential genetic engineering products: they range from moderately positive to overwhelmingly positive. At the top are "A treatment that would save the lives of people who have blood cancer" (89 points out of 100), "A genetically engineered drug that lowers blood pressure better than other drugs, reducing the risk of heart attack" (87 points), and "Genetically engineered cotton that resists insect pests -- this could greatly reduce the use of chemical pesticides" (87 points). A bit less popular, although still highly desirable in the public mind are "Healthier cooking oil and margarine, with more of the desirable, unsaturated fats and fewer of the undesirable fats" (79 points out of 100) "Genetically modified viruses to protect farm crops by attacking insect pests, such as beetles and locusts" (76 points), and "Modified viruses to control imported animal pests (such as rabbits or feral pigs) by preventing them from breeding" (76 points). The populace was a little less favorable towards "Leaner, healthier pork (assuming it is clearly labeled, so you can decide for yourself whether or not to buy it)", granting it a desirability rating of 73 points. There is still a substantial majority support for the modified pork, with 73% thinking it a "Good idea" or a "Very good idea". Support is lowest for the genetically engineered tomato: Australians accord it a desirability rating of 67 points out of 100. Even in this case, a clear majority are in favour, with 64% declaring genetically engineered tomatoes to be a "Good idea" or a "Very good idea".

It is clear the Australian public is broadly supportive of a wide range of genetic engineering projects. The average level of support is 79 points out of 100 -- so the average Australian rates the average genetic engineering project just a shade more favorable than a "good idea".

5.3.5 Potential Bias in These Questions?

It is clear that there is substantial variation in the public's views about different genetically engineered products. As in overseas studies, support is overwhelming for medical uses, high for general agricultural uses, and least for genetically modified foods. As a consequence, there is no entirely unambiguous answer to the question "how supportive of genetic engineering is the Australian public" -- it depends, in part, on which product you have in mind.

This ambiguity is quite normal for public policy issues. For example, there is no unambiguous answer to the question "how supportive of government spending is the Australian public?". Instead the answer depends on what the spending is for -- for example support is high for spending on education, moderate for spending on unemployment benefits, and low for spending on foreign aid.

Because of this variability, someone who wanted to paint an overly optimistic picture of public reaction to genetic engineering could concentrate entirely on medical uses and (correctly) report overwhelming support. But someone who wished to paint an overly pessimistic picture could concentrate entirely on food for human consumption, and (correctly) report only majority support, with a substantial minority uncertain. And in the extreme, someone could imagine a genetic engineering project with dubious goals, ask about it in a survey, and report widespread opposition.

We have chosen a middle road, reporting the average for a range of realistic projects which are already underway in Australia or overseas, eschewing the wild dreams of genetic engineering visionaries but also the dark scenarios of genetic engineering's most imaginative foes.

There is, however, one way in which our choice of questions has perhaps slightly tipped the scales against genetic engineering: our principal example, the genetically engineered tomato, is the least popular product on our list, and so, if anything, gives a slightly unfavourable introduction to the topic. A more neutral choice would have been one of the agricultural products , and a more favorable choice would be one of the medical products. We thought the simplicity and familiarity of tomatoes (and the fact that they are among the first genetically engineered products to reach the market overseas) outweighed this slight disadvantage.

5.4 Structure of Opinion on Genetic Engineering

A factor analysis shows that people who favour one genetic engineering product tend strongly to favour all of them, and conversely those who are dubious about one tend to be dubious about all. This justifies combining them into a single scale for subsequent analysis:

[Definition] Attitudes to Genetic Engineering = mean( Good vs. bad rating for: cancer treatment, blood pressure medicine, pest resistant cotton, viruses to protect crops, viruses against animal pests, new tomatoes, leaner pork, cooking oil with unsaturated fats)

There is some evidence of modest differences among these, particularly between medical products on the one hand and agricultural and food products on the other. But in the interest of simplicity, we ignore these differences here.

5.5 Summary: Approval

The Australian public is broadly supportive of a wide range of genetic engineering projects. The average Australian rates the average genetic engineering project as a "good idea".

The most popular genetic engineering products are a treatment for blood cancer, a drug that lowers blood pressure, and cotton that resists insect pests. More than 90% of Australians favour these. Then comes healthier cooking oil, genetically modified viruses to protect farm crops by attacking insect pests, and viruses to control imported animal pests and lean pork. Support is lowest for the genetically engineered tomato but even here a clear majority is in favour, 64% declaring them to be a "good idea" or a "very good idea" so long as they are clearly labeled. 

To top of page
To next chapter
Table of Contents