Consumer attitudes towards genetically modified foods

Appetite (2002) 39, 9±24 doi:10.1006/appe.2002.0486, available online at http://www.idealibrary.com on

1

Research Report

Consumer attitudes towards genetically modified foods Maria K. Magnusson and Ulla-Kaisa Koivisto Hursti Department of Public Health and Caring Sciences, Section of Caring Sciences,University of Uppsala,Uppsala Science Park, S-75183 Uppsala, Sweden (Received 26 September 2001, finalrevision 4 April 2002, accepted in revised form 5 April 2002)

The present study reports attitudes towards genetically modified (GM) foods among Swedish consumers. A random nation-wide sample of 2000 addressees, aged 18±65 years, were mailed a questionnaire and 786 (39%) responded. Most of these consumers were rather negative about GM foods. However, males, younger respondents and those with higher level of education were more positive than were females, older respondents and those with lower level of education. A majority of the consumers had moral and ethical doubts about eating GM foods and did not perceive attributes like better taste or lower price beneficial enough to persuade them to purchase GM foods. However, tangible benefits, like being better for the environment or healthier, seemed to increase willingness to purchase GM foods. # 2002 Elsevier Science Ltd. All rights reserved.

Introduction Consumer acceptance of genetic engineering (GE) has been found to vary between different types of applications of the technology (Fjñstad et al., 1998; Frewer & Shepherd, 1995; Gaskell et al., 2000), but also within the same category of application (Frewer et al., 1997). Consumers have been shown to hold more positive attitudes towards the use of GE for medical purposes than for food production purposes (Fjñstad et al., 1998; Frewer & Shepherd, 1995; Gaskell et al., 2000). In general, medical applications are perceived to be more beneficial, less risky and more ethically correct than food applications (Frewer & Shepherd, 1995). There is also evidence for that consumer acceptance depends on the type of organism being manipulated (Frewer & Shepherd, 1995; Frewer et al., 1997; Sparks et al., 1994). Frewer and Shepherd (1995) found that GE of microorganisms and plants was This research was supported with grants from the Swedish Council for Forestry and Agricultural Research. Address correspondence to: Maria K. Magnusson. Tel.: ‡46 18 471 6313; Fax: ‡46 18 471 3490. E-mail: [email protected] 0195±6663/02/$ ± see front matter

associated with less risk than GE of animals or human genetic material. In a British study, applications involving micoorganisms and plants were perceived to be beneficial, necessary and advantageous, whereas applications involving human DNA and animals were stated to be unethical, harmful and dangerous (Frewer et al., 1997). These findings stress the importance of studying attitudes to specific applications instead of attitudes towards the technology in general. Attitudes towards GE have been found to vary between countries (Bredahl, 1999; Gaskell et al., 1998). American and Canadian consumers seem to be more positive towards the technology (Pollara and Earnscliffe Research, 2000) than consumers in Europe (Fjñstad et al., 1998; Gaskell et al., 1998). In the third Eurobarometer report the Swedish consumers appeared to be more negative to GE than many other European consumers (Fjñstad et al., 1998; Gaskell et al., 1998). The fourth Eurobarometer survey showed that they still are more negative than consumers in many other European countries are (Gaskell et al., 2000). Also, a number of other demographic differences in consumer attitudes have been found. Research has shown that women tend to be more sceptical (Hoban, 1996) and # 2002 Elsevier Science Ltd. All rights reserved.

10

M. K. Magnusson and U.-K. K. Hursti

have more negative attitudes towards GE than men (Fjñstad et al., 1998; Gaskell et al., 1998; Hoban & Kendall, 1993; Koivisto Hursti et al., in press; Olofsson & Olsson, 1996). Age differences have also been suggested (Fjñstad et al., 1998; Koivisto Hursti et al., in press; Olofsson & Olsson, 1996) but the results have been contradictory. Some studies have shown that older subjects are the most positive ones (Fjñstad et al., 1998; Olofsson & Olsson, 1996) while others have revealed the opposite, younger being more positive (Koivisto Hursti et al., in press; Sparks et al., 1994). European consumers have shown rather negative attitudes towards employing GE in the area of food production (Gaskell et al., 1998, 2000; Grove-White et al., 1997). The fourth Eurobarometer survey demonstrated a general increase in the opposition towards genetically modified (GM) foods among European consumers from 1996 to 1999 (Gaskell et al., 2000). Research has found that consumers question the need of GE (Grove-White et al., 1997) and perceive that there is no need to use this technology in food production at all (Bredahl, 1999). Further, consumers seem to be of the opinion that ingestion of GM foods may reduce personal healthiness and can not be trusted because of its perceived unknown long-term consequences on human health and the environment (Bredahl, 1999). Canadian consumers have in general shown positive attitudes towards biotechnology but results from recent research indicate that of all the biotechnology areas, the food application is the one most capable of being redefined negatively (Pollara and Earnscliffe Research, 2000). Like the acceptance of GE in general the acceptance of GM foods has been found to depend on several factors such as tangible benefits of specific products (Frewer et al., 1996), moral and ethical considerations (Bredahl, 1999; Fjñstad et al., 1998; Gaskell et al., 1998), and perceived risk (Frewer et al., 1997; Sparks et al., 1994). Another factor of relevance is perceived control over the production of GM foods (Frewer et al., 1996). Findings from a Swedish focus group study suggest that the subjects' fear of GM is related to their fear of loosing control over their own lives (Wibeck, 1998). A number of consumers in the British focus group study expected that they already and unwittingly were eating GM foods (Grove-White et al., 1997). In the Canadian study, most consumers were very surprised to find out that GM ingredients are pervasive in processed foods (Pollara and Earnscliffe Research, 2000). The Canadian consumers showed an interest in more information on GM foods in order to facilitate informed choices (Pollara and Earnscliffe Research, 2000). Other studies have also shown that most consumers want GM foods to be labelled (FDA, 2000; Fjñstad et al., 1998; Gaskell et al., 1998; Hill et al., 1998; Koivisto Hursti et al., in press).

It has been suggested that the low acceptance of GM foods may be due to that most modifications so far have only profited the producers. Thus, it is reasonable to think that GM foods with tangible benefits may increase consumer acceptance. However, results regarding consumer willingness to purchase GM foods with tangible benefits seem to be ambiguous. Some studies suggest that benefits like ``more nutritious'', ``better for the environment'' (Frewer et al., 1996) and ``cheaper price'' (Grove-White et al., 1997) may influence consumer willingness to purchase GM foods. Nevertheless, a recent Nordic study demonstrated that the conventional product (cheese, candy and salmon) was the most preferred product although most of the GM alternatives possessed either benefits to human health, the environment or had a lower price (Grunert et al., 2001). In a Swedish survey, around one-fifth of the consumers were willing to purchase a GM tomato with better taste, better nutritional quality, longer shelflife or a lower price (Koivisto Hursti et al., in press). However, it is likely that these tangible benefits did not influence the willingness to buy the tomato since the same number of consumers stated that they would be interested in buying a GM tomato with no such benefits (Koivisto Hursti et al., in press). The overall objective of the present study was to gain further understanding about Swedish consumers' perceptions of GM foods. More specifically, the aims were to investigate consumer attitudes towards nine food applications of GE, attitudes towards GM foods in general, to what extent consumers are interested in buying GM foods with tangible benefits, and their knowledge about biology and genetics. Also, possible gender, age and educational differences were investigated. Further, we wanted to study possible relationships between the consumers' interest for natural foods, light food products and healthy eating (the Health Scales from Health and Taste Attitudes Scales, Roininen et al., 1999), and attitudes towards GM foods.

Method Respondents Two-thousand addressees aged between 18 and 65 years (this age range represents approximately 60% of the Swedish population) were recruited by random selection from the national population register (Sema Group, 2000). Questionnaires were mailed during May and June 2000, and addressees not responding were sent one reminder. Below, the most recent data available for the entire Swedish population (Statistics Sweden, 2001a,b) are given after the

Attitudes to GM foods 11

abbreviation Pop. The total number of respondents was 786 (39%), and of those were 48% men (Pop 49%) and 52% women (Pop 51%). Mean age was 41
9  13
6 years (Pop 40
2 years). The participants were quite evenly distributed across age groups: 18±25 years (15%; Pop 15%), 26±35 years (21%; Pop 23%), 36±45 years (21%; Pop 22%), 46±55 years (21%; Pop 23%), and 56±65 years (21%; Pop 18%). The majority were married or cohabitants (70%) and 26% were single. Twenty per cent had elementary school or 9 years of compulsory education (Pop 27%), 24% a degree from folk high-school or 2 years of upper secondary school (Pop 28%), 25% 3 years of upper secondary school (Pop 17%), and 30% university or university college education (Pop 26%). Thus, there is an over-representation of those with 3 years of upper secondary school and university education. Those who completed the questionnaire received a lottery ticket (value approximately US$ 2
50) or could donate the same amount to charities (Swedish Children's Cancer Foundation).

Questionnaire The questionnaire comprised questions concerning attitudes towards the use of GE in nine different food applications, attitudes towards GM foods in general and interest in purchasing GM foods with tangible benefits, knowledge about biology and genetics, and attitudes towards food and health (The Health Scales from Health and Taste Attitude Scales; Roininen et al., 1999). At the outset, the questionnaire stated the following definition of genetic modification. Genetic modification of plants and animals is done by means of genetic engineering by cutting a piece of DNA from one organism and transferring it into another in order to change its properties. For animals this is often done by injecting genes into cells, while for plants genes are usually shot or transferred by means of special bacteria. Classical breeding, e.g. crossing plants or inseminating animals is not considered as genetic modification even if this technique also aims at choosing superior genetic traits.

Questions Attitudes towards use of genetic engineering in nine food applications. Subjects were asked to rate nine applications of GE in food production on basis of thirteen different constructs on six point unipolar scales (Table 1). The food applications were ``Soy that has been GM to become resistant against pesticides'', ``Tomatoes that have been GM to stay fresh longer'', ``Fat-free yoghurt that has been produced with GM lactic acid bacteria'', ``Beer brewed with GM yeast'', ``Strawberries that have

been GM so that they can grow under dry conditions'', ``Pigs that have been GM in order to produce meat with lower fat content'', ``Wheat that has been GM to give higher yields'', ``Salmon that has been GM to grow ten times faster than `normal' salmon'', and ``Rice with a higher iron and b-carotene content which could help people in developing countries to meet their daily needs for those substances''. The constructs. The constructs employed in the questionnaire were elicited through interviews according to the repertory grid method (RGM; Kelly, 1955). The method has been employed earlier to investigate what terminology the public use when they express concerns about different applications of GE (Frewer et al., 1997). Forty-one subjects (14 men and 27 women) aged between 18 and 60 years from the Uppsala area were recruited to participate in an interview and questionnaire study. The interviews were conducted according to the RGM in a manner similar to the one employed by Frewer et al. (1997). Fifteen possible applications from the area of food production, agriculture and medicine were used. Half of the sample was interviewed about applications phrased in general terms and the other half about 15 more specific applications. All the subjects were given a standardised definition of GM before the start of the interview (the same as the one given in the questionnaire). Subjects were presented a set of three applications of GM in a randomised order. Each application was presented twice during the interview, giving 10 different combinations of applications in all. For each set of three applications the interviewees were asked ``Which of these applications gives you the most concern, and why?'', ``Which of these applications is the easiest for you to accept, and why?'', and ``Which of these applications is most difficult for you to accept, and why?''. All interviews were tape recorded and transcribed verbatim. The constructs that most subjects used when they expressed their concern/acceptance were employed in the present study (Table 1). Results from the interviews will be presented elsewhere. Attitudes towards GM foods with and without tangible benefits. Subjects were asked to estimate on a seven point bipolar scale (endpoints: disagree strongly, agree strongly, midpoint: neither agree nor disagree) the extent of their agreement with seven statements. Three of the statements were developed by Saba et al. (2000) and has been employed in a study among Italian consumers. The statements were ``I would feel guilty if I consumed foods derived from GE'', ``Consuming foods derived from GE goes against my principles'', ``It would be morally wrong for me to consume foods derived from GE'' (Saba et al., 2000), ``I would purchase GM foods if they (1) tasted better, (2) were

12

Construct

Abbreviation

End point ˆ 1

End point ˆ 6

1. How much benefit do you think that we can have from the following applications of genetic engineering? 2. How unethical do you perceive the following applications of genetic engineering? 3. How healthy do you think it is to eat the following foods produced by means of genetic engineering? 4. How great control (possibility to decide for yourself) over the consumption of the following foods produced by means of genetic engineering do you think you have? 5. How reluctant do you feel for the following applications of genetic engineering? 6. How strong concern do you feel for the following applications of genetic engineering? 7. To what extend do you think that we tamper with nature in the following applications of genetic engineering? 8. How great a risk do you perceive there is with using the following applications of genetic engineering? 9. How great a risk for misuse do you think there is with the following applications of genetic engineering? 10. To what extent do you perceive that the following applications of genetic engineering are used for profit alone? 11. To what extent do you think the following applications of genetic engineering serve a good purpose? 12. How necessary do you think the following applications of genetic engineering are? 13. To what extent do you think we have knowledge about the consequences of the use of the following applications of genetic engineering?

Benefit

No benefit at all

Very much benefit

Unethical

Not at all unetichal

Very unethical

Healthy

Not at all healthy

Very healthy

Control over the consumption

No control at all

Great control

Reluctance

No reluctance at all

Very strong reluctance

Concern

No concern at all

Very strong concern

Tampering with nature Risk

No tampering at all with nature No risk at all

To a great extent tampering with nature Great risk

Risk for misuse Used for profit alone Serves a good purpose Necessary

No risk at all for misuse

Great risk for misuse

Not used for profit at all No good purpose at all Not at all necessary

To a great extent used for profit alone Serves to a great extent a good purpose Very necessary

Knowledge about the consequences

No knowledge at all about the consequences

Very much knowledge about the consequences

M. K. Magnusson and U.-K. K. Hursti

Table 1. The constructs on which the nine food applications were rated

Attitudes to GM foods 13

healthier than conventional foods, (3) contributed to improve the general state of the environment, and (4) were cheaper''. Knowledge about biology and genetics. Knowledge was assessed with five statements (see Table 3). The statements ``yeast used for brewing beer consist of living organisms'', ``GM animals are always bigger than ordinary ones'', and ``more than half of human genes are identical for those of chimpanzees'' were employed in the questionnaire in the third Eurobarometer (Durant et al., 1998). Subjects marked with a cross whether they thought the statement was true, false or if they did not know. The Health and Taste Attitude Scales. A Swedish version of the Health Attitude Scales developed by Roininen et al. (1999) was employed to assess attitudes to food and health (see Table 7 for items). Subjects rated their extent of agreement on a seven point bipolar scale (endpoints: disagree strongly, agree strongly, midpoint: neither agree nor disagree).

post hoc comparisons. Bonferroni correction was utilised to adjust the level of significance for all t-tests and ANOVAs. To obtain an overall alpha level of 0
05 the alpha for each t-test/ANOVA had to be 0
00017 (SISA, 2002). Thus, the level of significance was set to p < 0
0001 for all t-tests and ANOVAs. Principal component factor analysis was performed on the 20 items in the Health Attitude Scale (Table 7). Briefly, factor analysis groups variables that vary together. The first factor is the single best combination of variables which accounts for the largest proportion of the total variance, the second factor represents the best combination accounting for the greatest proportion of the residual variance, and third and higher factors account for successively smaller proportions.

Subjects not responding to the original questionnaire

Attitudes towards the use of GE in nine food applications. In general, most subjects were rather negative towards the use of GE in food production (Table 2). The rice application was perceived to be the most positive one (highest benefit, least unethical, most healthy etc.; Table 2). In several cases, the applications concerning pork meat and salmon were the ones that the subjects were most negative towards (most unethical, most concern, highest risk etc.; Table 2). The figures for soy, tomatoes, yoghurt, beer, strawberries and wheat were fairly equal, but in some cases the wheat was perceived as more positive and beer as more negative than the other applications (Table 2). Attitudes towards GM foods with and without tangible benefits. The majority of subjects (62%) stated that it would be against their principles to consume GM foods and that it would be morally wrong for them to eat GM foods (58%). More than half of the subjects said that they would feel guilty if they ate GM foods (53%). Only between 12% and 13% said that they would be interested in buying GM foods even if they had lower price and tasted better. However, if they were healthier and better for the environment a substantially larger proportion of the subjects stated that they would buy such foods (31% and 43%, respectively). These percentages reflect the proportion of subjects who agreed or strongly agreed with the statements. Knowledge about biology and genetics. Most of the subjects had good knowledge about biology and genetics. Between 52% and 94% answered the five questions correctly (Table 3). The statement that most subjects had least knowledge about was ``GM animals are always bigger than `conventional' animals''. More than

Since the response rate was less than 50%, we wanted to look closer into the group of subjects not responding. A short questionnaire consisting of questions about attitudes towards GM foods in general and the demographics was sent to half the subjects (n ˆ 618) not responding to the original questionnaire and 104 (17%) of these responded. Also, 10% (n ˆ 61) of the other half of the subjects were randomly picked out and their telephone numbers were sought up in telephone directories. The subjects were phoned and asked if they would like to participate in a short telephone interview. Twenty-four (39%) agreed to participate. The questions posed in the telephone interviews were the same as those in the short questionnaire. There were no significant differences between those who filled out the short questionnaire/ were telephone interviewed and those who returned the original questionnaire concerning the demographic variables (age, gender, civil status and education). The most frequently given reasons for not answering the questionnaire were ``did not have time'' (37%) and ``a difficult topic'' (15%). Some also thought that the questionnaire contained too many questions.

Statistical methods Descriptive statistics, two-tailed t-test (unpaired), oneway analysis of variance (ANOVA), principal component factor analysis, and chi-square tests were used for the statistical analysis. Fisher's PLSD test was used for

Results Descriptive results

14

M. K. Magnusson and U.-K. K. Hursti

Table 2. Proportions (%) of consumers rating the levels of thirteen constructs resulting from the application of GM to nine foods Constructs (abbreviated) 1. Benefit No/very little Little/rather high High/very high 2. Unethical Not at all/very little Little/rather Unethical/very unethical 3. Healthy Not at all/very little Little/rather Healthy/very healthy 4. Control over the consumption No/very little Little/rather much Much/very much 5. Reluctance No/very little Little/rather strong Strong/very strong 6. Concern No/very little Little/rather strong Strong/very strong 7. Tampering with nature No/to a very little extent To a little/rather great extent To a high/very high extent 8. Risk No/very little Little/rather high High/very high 9. Risk for misuse No/very little risk Little/rather large Very large/great 10. Used for profit alone Not used/to a very little extent To a little/rather high extent To a high/very high extent 11. Serves a good purpose No/to a very little extent A little/rather high extent To a high/very high extent 12. Necessary Not at all/very little A little/rather Necessary/very necessary 13. Knowledge about the consequences No/very little Little/rather much Much/very much

Soy Tomatoes Yoghurt Beer Strawberries Pork meat Wheat Salmon Rice 61 33 7

64 29 6

69 26 6

81 16 3

64 29 7

75 20 5

55 31 15

82 15 3

25 33 42

23 32 45

23 27 50

24 27 50

22 23 55

24 27 48

11 18 71

29 27 44

12 14 74

46 31 23

68 25 7

72 22 7

68 22 10

76 19 5

70 23 7

76 16 8

64 26 10

84 11 5

37 36 27

75 18 6

70 23 8

68 23 9

74 18 7

71 21 8

70 21 9

78 16 6

73 18 9

74 19 7

15 34 51

13 29 58

14 29 57

14 27 59

15 32 54

7 17 76

20 29 51

6 15 79

36 35 29

16 38 46

15 36 49

16 36 48

19 33 48

17 34 49

7 23 70

19 33 48

5 22 73

30 38 33

7 32 61

5 29 67

7 33 60

9 32 59

6 29 65

4 16 80

8 29 62

3 14 83

10 37 53

10 41 49

10 37 52

11 40 49

14 38 48

12 38 50

5 24 72

12 39 49

3 21 75

17 42 41

10 34 56

9 31 60

10 36 54

12 35 53

11 31 57

4 21 75

11 32 57

4 20 76

16 37 46

7 29 64

4 20 76

6 34 60

6 28 66

4 22 74

4 19 78

5 20 75

3 10 87

25 39 35

59 34 7

67 28 5

66 29 5

81 16 3

63 31 6

77 18 5

56 29 14

84 12 4

24 32 44

72 23 4

79 19 2

80 18 2

89 9 1

76 22 2

85 14 1

63 29 8

91 8 1

31 37 32

73 24 3

74 22 3

74 23 3

75 21 3

74 24 2

77 20 3

73 24 3

79 18 3

67 28 4

Attitudes to GM foods 15

Table 3. Percentage of respondents who declared that the given statements were true or false or who did not know Conventional foods do not contain any genes Yeast used for brewing beer contains living organisms Genetically modified animals are always bigger than ``conventional'' animals More than half of the genes are identical for man and the chimpanzee All human cells contain DNA

True

False

Do not know

10 84* 29

67* 2 52*

23 14 19

66*

6

28

94*

1

5

* The true answer.

one-fourth (29%) stated that this was true. Between 5% and 28% declared that they did not know whether the statements were true or false. Almost all subjects (94%) knew that all human cells contain DNA.

Differences between demographic groups Statistically significant gender, age and educational differences were found with respect to the subjects' knowledge of and attitudes towards GM. Males, younger subjects and those with 3 years of upper secondary school education were significantly more positive towards the food applications than their comparison groups. However, the differences between the mean values were quite low. Since the mean ratings for the applications involving soy, tomatoes, yoghurt, beer, strawberries and wheat were rather equal, it was decided to compute a mean value of these applications to limit the total number of applications in the comparison analyses. These six foods will be referred to as vegetables, yoghurt and beer (VYB). Gender. Overall, males were significantly more positive towards the food applications than were females (Table 4). However, no statistically significant gender differences were found for the constructs ``control over the consumption'', ``used for profit alone'', ``serves a good purpose'', and ``knowledge about the consequences'' on any of the food applications. Also, no significant differences were found for the rice application. Female subjects stated to a significantly greater extent that it would be against their principles to eat GM foods and that they would feel guilty if they consumed GM foods than did the males (Table 4). Also, the females rated to a greater extent that it would be morally wrong for them to consume GM foods (Table 4). No significant difference was found for the willingness to purchase GM foods if they tasted better, were healthier, cheaper or contributed to improve the general state of the environment (Table 4). A larger proportion of the males knew that ordinary foods contain genes than did the females (2(2) ˆ 10
0, p < 0
01, fc ˆ 0
11). Further, females stated to a greater

extent that GM animals are always bigger than ordinary ones or did not know whether this was true or not as compared to the males (2 ˆ 22
1, df ˆ 2, p < 0
0001, fc ˆ 0
17). Education. In general, those with 3 years of upper secondary school were most positive towards the different applications in comparison to the other educational groups (Table 5). Subjects with elementary school education and 2 years of upper secondary school education appeared to be the most negative ones. Statistically significant differences between educational level and the ratings for nine of the constructs were found for rice (Table 5). For pork meat significant differences were found for the constructs ``benefit'' and ``healthy'', for VYB for ``benefit'', ``unethical'', ``healthy'', ``reluctance'', ``concern'', ``risk'', and ``risk for misuse'', and for salmon for ``benefit'' (Table 5). No statistically significant differences were found for any of the foods for the constructs ``control over the consumption'', ``tampering with nature'', and ``knowledge about the consequences'' (Table 5). No significant differences were found between educational level and willingness to purchase GM foods with or without tangible benefits (Table 5). The knowledge that ``ordinary'' foods contain genes (2(6) ˆ 48
4, p < 0
0001, fc ˆ 0
18), and that GM animals are not always bigger than ordinary one's (2 ˆ 74
7, df ˆ 6, p < 0
0001, fc ˆ 0
22) increased with level of education. Age Subjects in the youngest age group (18±25) were in general more positive towards the food applications than the remaining groups (Table 6). The older the subjects the more negative they were (Table 6). No statistically significant differences were found between the age groups and salmon and only one significant difference was found for pork meat (``control over the consumption'') where the younger subjects were more positive than the older ones (Table 6). Only one significant difference was found between age and willingness to purchase GM foods with or without tangible benefits (Table 6). The youngest age group (18±25) stated to a greater extent that they would

16

M. K. Magnusson and U.-K. K. Hursti

Table 4. Means and

S.D.s

of ratings by gender (only variables with significant differences; p < 0
0001) Men

Benefit Pork meat Salmon Unethical Pork meat Salmon Healthy Pork meat Salmon VYB Reluctance Pork meat Concern Pork meat Tampering with nature Pork meat Salmon VYB Risk Pork meat Risk for misuse Pork meat Necessary VYB Guilt Goes against my principles Morally wrong

Women

Gender differences

m

S.D.

m

S.D.

df

t

2
1 1
9

1
4 1
3

1
7 1
5

1
1 0
9

763 765

4
8 5
2

4
6 4
7

1
7 1
7

5
1 5
2

1
4 1
4

762 763

5
4 4
4

2
2 1
9 2
3

1
5 1
3 1
2

1
6 1
4 1
9

1
2 1
0 1
1

758 757 748

5
3 5
3 5
0

4
9

1
5

5
3

1
2

765

4
7

4
6

1
6

5
1

1
2

765

4
8

5
1 5
2 4
5

1
4 1
3 1
4

5
5 5
5 5
0

1
0 0
9 1
1

766 766 762

5
1 3
9 5
1

4
8

1
4

5
2

1
1

757

4
1

4
9

1
3

5
3

1
1

759

3
9

1
9 4
2 4
5

1
0 1
7 1
9

1
7 5
0 5
2

0
8 1
5 1
6

757 759 763

4
0 6
9 5
1

4
4

1
8

5
2

1
5

761

6
5

purchase GM foods if they were better for the environment than did those aged 36±65 years (Table 6). A larger proportion of the subjects in the age group 36±45 knew that ordinary foods contain genes as compared to those in the other age groups (2 ˆ 22
2, df ˆ 8, p < 0
01, fc ˆ 0
12). However, the knowledge about the fact that GM animals are not always bigger than ``ordinary'' animals decreased with increasing age (2 ˆ 41
1, df ˆ 8, p < 0
0001, fc ˆ 0
16).

Health attitude scales Principal component analysis. Principal component factor analysis, varimax rotation, orthogonal solution was performed of the 20 items in the Health Attitude Scales (Roininen et al., 1999). Item number 5 on the NPI scale was by accident not negatively worded in the Swedish version as it was in the original scale. However, after discussion with the originator (K. Roininen) it does not seem likely that this has

affected the results. A three-factor structure (eigenvalues > 1) was tested and approved. Fifty-five per cent of the variance was explained by the three factors. The factors were labelled with the same names as those used by Roininen et al. (1999). The first factor (Natural product interest; NPI) includes items regarding consumption of foods that do not contain additives and are unprocessed, the second factor (Light product interest; LPI) comprises statements related to an interest in eating reduced-fat foods, and the third (General health interest; GHI) contains items concerning healthy eating. Intercorrelations between the factors were 0
34 (NPI and LPI), 0
39 (NPI and GHI) and 0
12 (LPI and GHI). Reliability of the factors was examined in terms of homogeneity (Cronbach's a-coefficients; Table 7). The coefficients were 0
79 (NPI), 0
82 (LPI), and 0
88 (GHI). Two items belonging to the NPI scale (according to Roininen et al., 1999) ``In my opinion, organically grown foods are no better for my health than those grown conventionally'' and ``I do not care

Attitudes to GM foods 17

Table 5. Means and Fs of ratings by education (only variables with significant differences; Fs (3, 743±763), p < 0.0001. A ˆ elementary school/nine year compulsory education, B ˆ a degree from folk high-school/two years of upper secondary school, C ˆ Three years of upper secondary school, D ˆ university or university college education Groups Benefit Pork meat Salmon Rice VYB Unethical Rice VYB Healthy Pork meat VYB Reluctance Rice VYB Concern Rice VYB Risk Rice VYB Risk for misuse Rice VYB Used for profit alone Rice Serves a good purpose Rice Necessary Rice

F

A

B

C

D

1
7 1
5 3
5 1
9

1
6 1
4 3
6 1
9

2
2 1
9 4
4 2
6

1.9 1
8 3.9 2
2

7
3 8
7 9
5 16
5

3
4 4
2

3
3 4
4

2
6 3
5

3.0 4
1

8
5 10
6

1
7 1
9

1
6 1
8

2
2 2
4

1.9 2
1

8
4 10
5

3
8 4
7

3
6 4
7

2
9 4
0

3
3 4
4

9
3 9
0

3
9 4
4

3
9 4
5

3
1 3
7

3
5 4
1

9
8 10
7

4
2 4
5

4
3 4
6

3
5 3
9

4
0 4
3

10
2 11
6

4
4 4
7

4
6 4
9

3
7 4
1

4
1 4
5

11
9 14
4

4
3

3
9

3
2

3
8

13
1

3
4

3
8

4
4

4
0

10
3

3
0

3
4

4
0

3
4

10
5

about additives in my daily diet'' loaded higher on the GHI scale than on the NPI scale. Therefore, it was decided that for the further analysis these two items should be moved from the NPI scale to the GHI scale. There were no substantial changes in Cronbach's alphas of these two scales after the transfer of these two items (0
79, NPI and 0
87, GHI). In order to investigate if subjects who rated low or high, respectively, on the health scales differed in attitudes towards the applications of GE, subjects with a mean rating under 4 on each scale were classified as ``low'' NPI/LPI/GHI and those with a mean rating over 4 as ``high'' NPI/LPI/GHI. Comparisons were then performed between those who had low/high scores on the scales. No statistically significant differences were found for those who had low/high LPI scores. Low/high ratings on the Natural Product Interest scale. Those with a high NPI score were more negative towards

the applications (e.g. perceived that they were more unethical and felt greater reluctance and concern towards them) than were those with a low NPI score (Table 8). In accordance with this, those with a low NPI score perceived that the applications were more beneficial, healthier, necessary and served a good purpose to a greater extent than did those with a high NPI score (Table 8). No significant differences were found for the constructs ``control over the consumption'' and ``knowledge about the consequences''. Those with a high NPI score agreed to a lesser extent with statements ``I would purchase foods produced by GE if they tasted better, were healthier, more environmentally friendly, and cheaper than did those with low NPI scores (Table 8). Further, subjects who scored high on the NPI scale agreed to a greater extent with the statements ``I would feel guilty if I consumed foods derived from GE'', ``Consuming foods

18

M. K. Magnusson and U.-K. K. Hursti

Table 6. Means and Fs of ratings by age groups (only variables with significant differences), Fs (4, 744±764), p < 0.0001. I ˆ 18±25 years, II ˆ 26±35 years, III ˆ 36±45 years, IV ˆ 46±55 years, V ˆ 56±65 years Groups Benefit Rice VYB Unethical Rice VYB Healthy Rice VYB Control over the consumption Pork meat Reluctance Rice VYB Concern Rice VYB Tampering with nature VYB Risk Rice VYB Risk for misuse Rice VYB Used for profit alone Rice Serves a good purpose Rice Necessary Rice Improve the environment

F

I

II

III

IV

V

4
7 2
8

4
1 2
3

3
9 2
0

3
5 2
0

3
4 2
0

12
3 11
9

2
3 3
3

2
7 4
0

2
9 4
0

3
5 4
3

3
4 4
3

12
6 8
4

3
8 2
6

3
5 2
2

3
1 2
0

3
0 2
0

2
9 1
8

7
2 8
4

2
7

2.3

2.2

1
9

1
9

7
0

2
4 3
7

3
2 4
3

3
4 4
6

3
8 4
7

3
8 4
6

15
0 9
6

2
8 3
5

3
3 4
0

3
6 4
3

4
0 4
5

4
0 4
4

12
1 11
0

4
3

4
6

4
8

5
0

4.8

6
4

3
4 3
7

3
7 4
1

4
1 4
4

4
3 4
6

4
2 4
5

9
7 10
8

3
5 3
8

3
9 4
5

4
3 4
6

4
5 4
8

4
4 4
7

9
1 10
5

3
2

3
2

3
8

4
1

4
3

15
4

4
9

4
2

4
0

3
4

3
3

19
3

4
4 4
6

3
8 4.1

3
4 3
6

3
0 4
0

3
0 3
7

16
8 6
9

derived from GE goes against my principles'', and ``It would be morally wrong for me to consume foods derived from GE'' than did those with a low NPI score (Table 8). Low/high ratings on the General Health Interest scale. The pattern for the GHI scale is the same as for the NPI scale. Thus, those who had high scores on the GHI scale appeared to be more negative towards the applications than those scoring low on the GHI scale (Table 9). Statistically significant differences were found for all constructs except ``control over the consumption'' and ``knowledge about the consequences''. In accordance with the findings above those with a high GHI score agreed to a greater extent that they would feel guilty and that it would be against their

principles to consume GM foods than did those with a low GHI score (Table 9). Likewise, subjects with a high GHI score stated to a greater extent that it would be morally wrong for them to consume GM foods (Table 9). Those with a low GHI score were more willing to buy GM foods with better taste and if they were more environmentally friendly than were those with a high GHI score (Table 9).

Comparisons between subjects responding and not responding to the original questionnaire Attitudes towards GM foods with and without tangible benefits. No statistically significant differences were found between those who answered the short

Table 7. Mean ratings, standard deviations and factor loadings on the subscales of the health attitude scales (Roininen et al., 1999) Statements

Mean

Factor loadings

S.D.

NPI

* A negative statement, the scale has been recoded.

GHI

4
5 5
2 4
2 5
0 4
5 4
9

1
3 1
4 1
3 1
6 1
5 1
6

0
76 0
34 0
69 0
64 0
65 0
10

ÿ0
06 ÿ0
12 0
02 ÿ0
05 ÿ0
19 ÿ0
09

0
26 0
56 0
03 0
25 0
14 0
47

4
1 4
2 4
3 4
0 3
7 3
6

1
5 1
5 1
4 1
5 1
5 1
5

ÿ0
30 ÿ0
31 ÿ0
26 0
08 0
17 0
10

0
67 0
61 0
62 0
82 0
83 0
70

0
30 0
33 0
29 ÿ0
09 ÿ0
12 ÿ0
23

4
9 4
7 4
9 4
2 4
3 4
8 4
8 4
3

1
4 1
2 1
5 1
4 1
2 1
3 1
4 1
5

0
02 0
63 0
29 0
32 0
62 0
53 0
07 0
09

ÿ0
01 0
10 0
10 0
39 0
16 0
15 0
00 0
18

0
72 0
54 0
73 0
44 0
51 0
55 0
73 0
65

Attitudes to GM foods 19

Natural product interest 1. I try to eat foods that do not contain additives 2. *I do not care about additives in my daily diet 3. I do not eat processed foods, because I do not know what they contain 4. I would like to eat only organically grown vegetables 5. In my opinion, artificially sweetened foods are harmful for my health 6. *In my opinion, organically grown foods are no better for my health than those grown conventionally. Eigenvalue ˆ 5
9, Variance % ˆ 29
3, Cronbach's a ˆ 0
79 Light product interest 1. *I do not think that light products are healthier than conventional products 2. *In my opinion, the use of light products does not improve one's health 3. *In my opinion, light products don't help to drop cholesterol levels 4. I believe that eating light products keep one's cholesterol level under control 5. I believe that eating light products keeps one's body in good shape 6. In my opinion, by eating light products one can eat more without getting too many calories. Eigenvalue ˆ 3
5, Variance % ˆ 17
3, Cronbach's a ˆ 0
82 General health interest 1. *The healthiness of food has little impact on my food choices 2. I am very particular about the healthiness of food I eat 3. *I eat what I like and I do not worry much about the healthiness of food 4. It is important for me that my diet is low in fat 5. I always follow a healthy and balanced diet 6. It is important for me that my daily diet contains a lot of vitamins and minerals 7. *The healthiness of snacks makes no difference to me 8. *I do not avoid foods, even if they may raise my cholesterol. Eigenvalue ˆ 1
7, Variance % ˆ 8
7, Cronbach's a ˆ 0
88

LPI

Table 8. Means and S.D.s of ratings by respondents scoring high and low on the Natural Product Interest (NPI) scale (only variable with significant differences, p < 0
0001) Low NPI

High NPI

df

t

m

S.D.

m

S.D.

Benefit Pork meat Salmon Rice VYB

2
4 2
3 4
5 2
8

1
5 1
5 1
6 1
2

1
7 1
4 3
7 2
0

1
1 0
9 1
8 1
0

757 760 756 738

7
5 9
8 5
8 9
4

Unethical Pork meat Salmon Rice VYB

4
3 4
3 2
5 3
2

1
7 1
7 1
5 1
5

5
1 5
1 3
2 4
3

1
4 1
4 1
7 1
6

755 756 750 746

5
6 6
2 5
1 8
1

Healthy Pork meat Salmon Rice VYB

2
6 2
1 3
9 2
7

1
6 1
4 1
6 1
2

1
6 1
5 3
0 1
9

1
2 1
0 1
7 1
1

752 751 744 742

8
6 7
3 6
5 9
3

Reluctance Pork meat Salmon Rice VYB

4
3 4
5 2
5 3
4

1
6 1
6 1
5 1
5

5
4 5
5 3
7 4
8

1
2 1
0 1
7 1
3

758 759 748 754

9
5 9
3 8
5 12
1

Concern Pork meat Salmon Rice VYB

3
9 4
1 2
7 3
1

1
7 1
7 1
5 1
4

5
2 5
3 3
9 4
5

1
2 1
0 1
6 1
3

758 759 750 753

11
4 11
5 9
0 13
0

Tampering with nature Pork meat Salmon Rice VYB

4
8 4
9 3
9 4
1

1
4 1
4 1
5 1
3

5
4 5
5 4
6 4
9

1
0 1
0 1
4 1
2

759 759 751 755

6
2 6
0 6
0 8
0

Risk Pork meat Salmon Rice VYB

4
3 4
4 3
1 3
4

1
5 1
4 1
3 1
3

5
2 5
3 4
3 4
6

1
1 1
0 1
4 1
2

752 752 747 750

9
3 9
5 9
9 12
3

Risk for misuse Pork meat Salmon Rice VYB

4
5 4
6 3
3 3
7

1
4 1
5 1
5 1
4

5
3 5
3 4
4 4
8

1
0 1
0 1
5 1
2

753 754 745 749

8
3 8
1 8
6 10
2

Used for profit alone Pork meat Rice VYB

4
8 3
3 4
6

1
2 1
6 1
1

5
3 3
9 5
1

1
1 1
6 1
0

750 740 745

4
8 4
8 6
0

Serves a good purpose Pork meat Salmon Rice VYB

2
3 2
0 4
4 2
7

1
4 1
3 1
6 1
1

1
7 1
5 3
8 2
0

1
1 1
0 1
8 1
0

755 756 750 751

5
7 5
3 4
6 7
8

Necessary Pork meat Salmon Rice VYB

1
9 1
6 4
0 2
2

1
2 1
0 1
6 1
0

1
5 1
3 3
3 1
7

0
8 0
7 1
7 0
8

752 754 747 749

5
5 4
8 4
9 7
6

Guilt Goes against my principles Morally wrong Better taste Healthier Improve the environment Cheaper

3
4 3
5 3
6 3
4 4
1 4
6 3
4

1
7 1
8 1
7 1
7 1
7 1
6 1
7

5
0 5
3 5
2 2
2 3
1 3
7 2
3

1
5 1
5 1
5 1
4 1
8 1
8 1
5

752 756 754 757 757 755 758

12
2 13
7 12
4 10
0 6
7 5
8 8
3

Attitudes to GM foods

21

Table 9. Means and S.D.s of ratings by respondents scoring high and low on the General Health Interest (GHI) scale (only variables with significant differences, p < 0
0001) Low GHI Benefit Salmon VYB Unethical VYB Healthy Pork meat VYB Reluctance Salmon Rice VYB Concern Pork meat Salmon Rice VYB Tampering with nature VYB Risk Pork meat Salmon Rice VYB Risk for misuse Pork meat Salmon Rice VYB Used for profit alone Salmon VYB Serves a good purpose VYB Necessary Pork meat VYB Guilt Goes against my principles Morally wrong Better taste Cheaper

High GHI

df

t

m

S.D.

m

S.D.

2
0 2
6

1
3 1
1

1
6 2
1

1
0 1
1

755 735

4
4 5
5

3
5

1
5

4
2

1
6

742

4
8

2
3 2
5

1
5 1
2

1
8 2
0

1
3 1
1

750 740

4
0 5
7

4
8 2
8 3
7

1
5 1
6 1
5

5
3 3
5 4
6

1
2 1
8 1
4

756 745 752

4
8 4
7 6
7

4
3 4
4 2
9 3
4

1
6 1
6 1
5 1
4

5
0 5
2 3
7 4
4

1
3 1
2 1
7 1
4

756 756 747 751

5
6 6
4 5
9 7
9

4
3

1
3

4
8

1
3

753

4
8

4
6 4
7 3
3 3
7

1
5 1
4 1
5 1
4

5
1 5
2 4
2 4
5

1
1 1
1 1
5 1
3

750 749 744 749

4
6 5
0 6
6 6
7

4
7 4
7 3
4 3
9

1
4 1
4 1
6 1
4

5
2 5
3 4
4 4
7

1
1 1
1 1
5 1
3

752 752 743 749

4
7 5
1 7
3 6
9

5
2 4
6

1
2 1
1

5
6 5
1

0
9 1
0

749 742

4
2 4
5

2
5

1
1

2
1

1
0

748

4
4

1
8 2
1 3
8 3
9 3
8 3
2 3
2

1
1 0
9 1
6 1
9 1
7 1
7 1
7

1
5 1
7 4
8 5
2 5
0 2
3 2
4

0
9 0
8 1
6 1
7 1
6 1
5 1
6

750 747 749 753 752 754 755

4
2 4
8 7
5 8
5 8
3 5
9 5
8

questionnaire/were telephone interviewed and those responding to the original questionnaire.

Discussion In general, most consumers were rather negative towards the use of GE within the food production

area. However, there was a small group of consumers that was more positive to the use of GE in food production. Males, younger subjects and those with 3 years of upper secondary school education were significantly more positive towards the food applications than were women, older subjects and the other educational groups. The applications that received the most positive ratings among all subjects were ``rice with

22

M. K. Magnusson and U.-K. K. Hursti

a higher iron and b-carotene content'' and ``wheat that has been GM to give higher yields''. A majority declared that it would be against their principles, and that it would be morally wrong for them to eat GM foods. Our finding that consumers are generally rather negative towards the use of GE for food production purposes is in accordance with other European consumer surveys (Fjñstad et al., 1998; Frewer et al., 1997; Gaskell et al., 1998; Koivisto Hursti et al., in press). The third Eurobarometer report demonstrated that Sweden was one of nine countries where the public were against the use of GE in food production (Gaskell et al., 1998). Nevertheless, not all subjects in the present study were negative. Males demonstrated more positive attitudes towards GM foods than did women. This finding is in accordance with previous research (Gaskell et al., 1998; Hill et al., 1998; Hoban & Kendall, 1993; Koivisto Hursti et al., in press; Olofsson & Olsson, 1996; Sparks et al., 1994). Further, subjects with 3 years of upper secondary school education were most positive towards the food applications in comparison to the other three educational groups. Other studies have indicated that people with higher education hold more positive attitudes towards GE (Fjñstad et al., 1998; Gaskell et al., 1998; Hoban & Kendall, 1993). However, in a recent Swedish study no educational differences were found (Koivisto Hursti et al., in press). It has been argued that knowledge is an important determinant of support for science and technology, i.e. the more informed the public are, the more likely they are to be supportive (Gaskell et al., 1998). However, results from the third Eurobarometer report indicate that knowledge plays at best a modest part in determining attitudes towards science and technology (Gaskell et al., 1998). Gaskell et al. (1998) concluded that knowledge is one factor influencing the formation of opinions, but those opinions may be positive or negative. Most consumers in the present study appeared to have good knowledge about biology and genetics. Nevertheless, more than one-fourth (29%) thought that GM animals are always bigger than conventional animals, which is false, and 23% and 28%, respectively, declared that they did not know whether the statements ``conventional foods do not contain any genes'' and ``more than half of the genes are identical for man and the chimpanzee'' were true or false. A comparison between the figures (for Sweden) for the statements employed in the third Eurobarometer report (Gaskell et al., 1998) and in the present study imply that the level of knowledge has not changed much during the last years. In our study, subjects in the youngest age group (with some exceptions) were in general more positive

towards GM foods than were the older ones. This is in accordance with the results from a study by Koivisto Hursti et al. (in press). The third Eurobarometer report found that supporters of food biotechnology were significantly more likely to be younger (Gaskell et al., 1998), and a British survey showed that older subjects reported lower benefits of GE compared to younger subjects (Sparks et al., 1994). However, some studies have indicated a reversed pattern. Olofsson & Olsson (1996) found that the older the subjects the more positive they were towards the manipulation of the human genome. The Swedish results from the Eurobarometer report demonstrated a similar finding, older people being more positive about drug production by GM organisms and genetic testing (Fjñstad et al., 1998). These results indicate that the pattern of the age differences is dependent on the application area, indicating that older subjects may be more positive towards medical applications (Fjñstad et al., 1998; Olofsson & Olsson, 1996), and younger more positive towards food applications (Gaskell et al., 1998; Koivisto Hursti et al., in press). Thus, attitudes towards GM seem to be related to the individual's age and level of education. However, how age and educational level influence attitudes is an issue that needs further investigation. Previous research has shown that different factors, e.g. risk, benefit, and moral/ethical issues influence consumer acceptance of GM foods. A majority of the consumers in our study stated that it would be against their principles, and morally wrong for them to eat GM foods, and that they would feel guilty if they ate such foods. This is in accordance with several other studies that have shown that moral and ethical issues influence consumer acceptance of GM foods (e.g. Fjñstad et al., 1998). Saba et al. (2000) found that moral obligation was a moderately good predictor of future consumption of GM foods among Italian consumers. However, attitude (towards the use of gene technology in food production) seemed to be a more important predictor than moral obligation (Saba et al., 2000). In general, most of the food applications were perceived as negative by a majority of the subjects but there were differences between the ratings of the applications. The fast growing salmon and lean pork meat were often given the most negative ratings, and the rice and wheat the most positive ones. The positive perceptions of the rice and wheat may be a reflection of perceived benefit as both applications may help to feed the world's growing population. The rice application could also have the possibility to help to improve the nutritional status of people in developing countries who have problems meeting their needs of vitamins and minerals. Another explanation for the positive attitudes may be that applications involving plants are

Attitudes to GM foods

often perceived as more beneficial and advantageous (Frewer et al., 1997; Hill et al., 1998), and are associated with less risk (Frewer & Shepherd, 1995) than those involving animals. It has also been suggested that acceptance towards GM in food production would increase if GM foods possessed tangible benefits for the consumers. Most consumers in our study did not perceive the attributes ``better taste'' and ``lower price'' as sufficiently good arguments for purchasing GM foods. This finding is interesting, as these two characteristics are usually two very important food purchase criteria. It may be interpreted to reflect that for GM foods the production method is a characteristic, which is regarded as more important than other criteria otherwise considered as highly important. Grunert et al. (2001) demonstrated that consumers regard ``non-GM'' as a value in itself. A substantial group of consumers in our study stated that they would purchase GM foods if they were ``healthier'' or ``better for the environment''. This finding corresponds with results from a British study indicating that British consumers were more likely to buy GM tomatoes if they were more nutritious or if they were better for the environment (Frewer et al., 1996). However, results from Grunert et al. (2001) suggest that remote societal benefits (like benefits to the environment) are not a good promoter of GM acceptance. In order to investigate if there is a relationship between health attitudes and attitudes towards GM foods, the Health attitude scales developed by Roininen et al. (1999) were employed. Those who scored high on the NPI scale were in general more negative towards GM foods than those who had lower scores. Also, those who scored high on the GHI scale were more negative towards the food applications of GE in comparison to those who scored low on that scale. This may be interpreted to reflect that GM foods are not perceived as natural or healthy by a large number of consumers. This confirms the results from earlier studies that have shown that GE products are perceived as less natural than conventional counterparts (Frewer et al., 1996) and that GE of vegetables and animals is seen as unnatural (Hill et al., 1998). A study that compared consumer perceptions of conventional foods with GM foods demonstrated that the main advantage of the conventional foods was that they were not GM (Grunert et al., 2001). The property of not being GM was related to safety and good health (Grunert et al., 2001). Research has also suggested that ``healthy'' eating is described in terms of eating natural foods (Connors et al., 2001; Povey et al., 1998; Santich, 1994) and ``unhealthy'' eating in terms of eating manufactured or processed foods (Povey et al., 1998; Santich, 1994).

23

Methodological considerations The relatively low response rate (39%) may indicate that many consumers are not familiar with the topic or do not have a definite opinion on GM foods. Another contributory explanation could be the length of the questionnaire. However, it is more likely that lack of time and the topic explain the low response rate as the most frequently given answers for not responding were ``did not have time'' and ``a difficult topic''. We have performed questionnaire studies with the same methodology (sampling, reminders, reward) and size of questionnaire in previous samples of the Swedish population (cf. Magnusson et al., 2001). In these, however, the themes have dealt with more familiar topics (e.g. attitudes to milk or organic foods) and the response rate has been between 20% and 30% higher. Our sample represented the Swedish population well regarding gender and age, but contained an overrepresentation of those with higher education. No statistically significant differences, neither demographic nor attitudinal, between subjects who answered the original questionnaire and subjects who did not respond the original questionnaire were found. Therefore, we consider our sample to be well representative of the Swedish population, despite the relatively low response rate.

Conclusions The results suggest that although the majority of the Swedish consumers are rather negative towards GM foods some consumers are fairly positive (males, younger subjects, those with three years of upper secondary school education). A majority has moral and ethical doubts about eating GM foods and does not perceive attributes like better taste or lower price beneficial enough to persuade them to purchase GM foods. Tangible benefits like better for the environment or healthier seem to increase consumer willingness to purchase GM foods. The present study contributes to further understanding on consumers' perceptions of GM foods. However, motives underlying consumer responses to GM foods and the influence of demographics, especially age and education, need to be further investigated. Consumers' underlying motives are very complex and therefore difficult to study. Different kinds of interview techniques (both group and individual) should be employed in future studies as they can bring out further knowledge regarding beliefs and values related to GE and GM foods.

24

M. K. Magnusson and U.-K. K. Hursti

References Bredahl, L. (1999). Consumers' cognitions with regard to genetically modified foods. Results from a qualitative study in four countries. Appetite 33, 343±360. Connors, M., Bisogni, C. A., Sobal, J. & Devine, C. M. (2001). Managing values in personal food systems. Appetite 36, 189±200. Durant, J., Bauer, M. W. & Gaskell, G. (1998). Biotechnology in the Public Sphere. A European Sourcebook. London: Science Museum. Food and Drug Administration (FDA) (2000). Report on consumer focus groups on biotechnology. Available at http://www.cfsan.fda.gov/~comm/biorpt.html. Fjñstad, B., Olsson, S., Olofsson, A. & von BergmannWienberg, M.-L. (1998). Sweden. In J. Durant, M. W. Bauer & G. Gaskell (Eds), Biotechnology in the Public Sphere. AEuropean Sourcebook. London: Science Museum. Frewer, L. J. & Shepherd, R. (1995). Ethical concerns and risk perceptions associated with different applications of genetic engineering: Interrelationships with the perceived need for regulation of the technology. Agriculture and Human Values, 12, 48±57. Frewer, L. J., Howards, C. & Shepherd, R. (1996). The influence of realistic product exposure towards genetic engineering of food. Food Quality and Preference 7, 61±67. Frewer, L. J., Howards, C. & Shepherd, R. (1997). Public concerns in the United Kingdom about general and specific applications of genetic engineering: risk, benefits and ethics. Science, Technology and Human Values 22, 98±124. Gaskell, G., Bauer, M. W. & Durant, J. (1998). Public perceptions of biotechnology in 1996: Eurobarometer 46
1. In J. Durant, M. W. Bauer & G. Gaskell (Eds), Biotechnology in the Public Sphere. A European Sourcebook. London: Science Museum. Gaskell, G., Allum, N., Bauer, M., Durant, J., Allansdottir, A., Bonfadelli, H., Boy, D., de CheveigneÂ, S, Fjñstad, B., Gutteling, J. M., Hampel, J., Jelsùe, E., Jesuino, J. C., Kohring, M., Kronberger, N., Midden, C., Nielsen, T. H., Przestalski, A., Rusanen, T., Sakellaris, G., Torgersen, H., Twardowski, T. & Wagner, W. (2000). Biotechnology and the European public. Nature Biotechnology 18, 935±938. Grove-White, R., Macnaghten, P., Mayer, S. & Wynne, B. (1997). Uncertain World. Genetically Modified Organisms, Food and Public Attitudes in Britain. Lancaster: Lancaster University. Grunert, K. G., LaÈhteenmaÈki, L., Nielsen, N. A., Poulsen, J. B., Ueland, O. & AÊstroÈm, A. (2001). Consumer perceptions of foods products involving genetic modification ± results from a qualitative study in four Nordic countries. Food Quality and Preference, 12, 527±542.

Hill, R., Stanisstreet, M., Boyes, E. & O'Sullivan, H. (1998). Reactions to a new technology: students' ideas about genetically engineered foodstuffs. Research in Science and Technological Education 16, 203±216. Hoban, T. J. & Kendall, P. A. (1993). Consumer Attitudes about Food Biotechnology. North Carolina: North Carolina State University. Hoban, T. J. (1996). Anticipating public reaction to the use of genetic engineering in infant nutrition. American Journal of Clinical Nutrition 63, 657±662. Kelly, G. A. (1955). The Psychology of Personal Constructs: A Theory of Personality. New York: Norton. Koivisto Hursti, U.-K., Magnusson, M. K. & Algers, A. Swedish consumers' opinions about gene technology. British Food Journal (In press). Magnusson, M. K., Arvola, A., Koivisto Hursti, U.-K., AÊberg, L. & SjoÈdeÂn, P.-O. (2001). Attitudes towards organic foods among Swedish consumers. British Food Journal 103, 209±226. Olofsson, A. & Olsson, S. (1996). The new biotechnology: Media coverage and public opinion. In B. Fjñstad (Ed.), Public Perceptions of Science, Biotechnology and a New University. Report 1996:10, Mid Sweden University. Pollara and Earnscliffe Research. (2000). Public opinion research into biotechnology issues ± overall narrative. Available at http://strategis.ic.gc.ca/SSG/bh00257e.html. Povey, R., Conner, M., Sparks, P., James, R. & Shepherd, R. (1998). Interpretations of healthy and unhealthy eating, and implications for dietary change. Health Education Research 13, 171±183. Roininen, K., LaÈhteenmaÈki, L. & Tuorila, H. (1999). Quantification of consumer attitudes to health and hedonic characteristics of foods. Appetite 33, 71±88. Saba, A., Rosati, S. & Vassallo, M. (2000). Biotechnology in agriculture. Perceived risks, benefits and attitudes in Italy. British Food Journal 102(2), 114±121. Santich, B. (1994). Good for you: beliefs about food and their relation to eating habits. Australian Journal of Nutrition and Dietetics 51, 68±73. Sema Group (2000). Infodata DAFA AB, Box 34101, S-100 26 Stockholm, Sweden. Simple Interactive Statistical Analysis (SISA) (2002). Available at: http://home.clara.net/sisa/bonhlp.htm#Input. Sparks, P., Shepherd, R. & Frewer, L. J. (1994). Gene technology, food production, and public opinion: A UK study. Agriculture and Human Values, 11, 19±28. Statistics Sweden (2001a). Available at: http://www.scb.se/ befovalfard/befolkning/befstor/befarlig/befarlig31tab.asp. Statistics Sweden (2001b). Available at: http://www.scb.se/ utbildning/bef_utbild/utbreg/utbregtab.asp. Wibeck, V. (1998). FoÈrestaÈllningar om genmodifierade livsmedel. LinkoÈpings universitet: Tema K, LinkoÈping.