ATTITUDES TO GENE TECHNOLOGY: THE SIGNIFICANCE OF TRUST IN INSTITUTIONS

ABSTRACT

This is a study of the relationship between trust in institutions and attitudes to gene technology in general, and GM food and stem cell research in particular. The role of so-called active trust is emphasised, meaning that trust is neither conceived as a trait nor a one-dimensional concept. The study uses data from a Eurobarometer survey of gene technology in Europe, conducted in 2002. People's attitudes in five European counties, France, Germany, Italy, Sweden and United Kingdom are compared, and the significance of trust in institutions in these countries is investigated. The results show that trust in institutions has an impact on attitudes to gene technology. Trust in experts, stakeholders and official bodies are associated with positive attitudes to GM food and stem cell research, whereas trust in Non-Governmental Organisations is associated with negative perceptions of these technologies. This confirms the significant role of active trust.

Keywords:

• institutional trust
• active trust
• attitudes to gene technology
• Eurobarometer
• European public

This article is based on the findings from a project ‘Life Sciences in European Society’ funded by the European Commission Directorate General for Research (QLG7-CT-1999-00286).

Notes

¹We define institutions in line with Sztompka's (1999: 43–4) definition. He writes: ‘… the trust directed at institutions and organizations (understood as specific structural arrangements within which actions and interactions take place). The school, the university, the army, the church, the courts, the police, the banks, the stock exchange, the government, the parliament, the industrial enterprise, and so forth, are typical objects for this type of trust’. This is comparable to the definition given in the Oxford Advanced Learner's Dictionary: ‘[an institution is] an organization established for social, educational, religious, etc. purposes’, i.e., influential societal organisations such as the government, mass media, universities, and industry.

²For a survey of the nature of attitudes and public opinion, see e.g., Oskamp (1991), Ajzen and Fishbein, (1980), Heijs and Midden (1997).

³The choice of applications is based on earlier research which shows that people perceive ‘green’, i.e., agricultural, and ‘red’ i.e., medical, gene technology, differently (see Bauer and Gaskell 2001; Durant et al. 1998; Hampel et al. 2000).

⁴In the literature trust is separated from confidence, where the former is associated with risk and vulnerability, and the latter with familiarity and earlier experiences (Siegrist et al. 2005: 147).

⁵Psychologists working from personality theory often view trust as a trait that individuals develop in varying degrees, and the focus is on individual and group differences across time. The most frequent method is psychometric scaling, and many other scholars have used scales developed by psychologists also to measure generalized trust (cf. Slovic 1999). Another common conceptualisation of trust is operationalised by game theory known as the ‘prisoner's dilemma’ which has been used by behavioural psychologists, political scientists and economists.

⁶The relationship between trust and attitudes has been empirically investigated by psychologists (for an overview see Viklund 2003). Trust is however most often defined as interpersonal trust and trust as a trait. Sociological studies of trust have primarily focused on cooperation in social and political settings, trust as a collective attribute and trust as social capital (Misztal, 1996; Putnam, 1993). More recently trust has also been studied in surveys with the aim of finding a relationship between trust and stable democracies (Inglehart 1999), and trust in institutions has been in focus in studies of welfare development and food safety (e.g., Berg et al. 2005;Kohring et al. 2006; Svallfors 2002). For a thorough survey and testing of the origins of individual trust, see Delhey and Newton (2003).

⁷It is important to point out that distrust, such as suspicion, monitoring and activation of institutional safeguards, is also a way of reducing complexity.

⁸The behavioural dimension is the action which the trust in a person or institutions results in, and as Lewis and Weigart (1985), we will not discuss this dimension any further.

⁹In this study we do not differentiate between trust in the institution per se, or in the reputation of the institution.

¹⁰Luhmann (1999) does not see trust as an individual trait or rationale at the individual level; trust is rational and functional for maintaining the system, i.e., he discards the rational choice perspective. Rational choice theory views trust as a purposive behaviour aiming at maximization of utility under risk, see e.g., Coleman (1990). For a critical evaluation of the use of rational choice theory and trust see Sato (2002).

¹¹Wynne (1996) criticises Giddens’ conceptualisation of trust, which according to Wynne develops from an automatic (or abstract) trust during early modernity to a reflexive (or active) trust during late modernity. Wynne argues that non-reflexive trust has never prevailed among the public, and that the active trust concept is based on a rational choice assumption, where the individual chooses who to trust in a specific situation. Wynne discards this idea for a more hermeneutic assumption, where the public responses to, for example, scientific expertise are based on an awareness of and accommodation to social dependence on expert institutions (Wynne 1996). The concept of active trust in relation to interpersonal relationships has also been criticised (see e.g., Adams 2004; Mestrovic 1998). The main argument against active trust in personal relationships is that that it per se cannot replace e.g., faith, because it would mean constan t questioning, for instance a partner's faithfulness and honesty, indicating the opposite of trust.

¹²Germany was divided in two parts, East and West, each with a sample of 1,000 people.

¹³The knowledge variable is not an absolute measurement of knowledge but an indicator of people's knowledge (if they know genes exist in all living things etc.), a way of discriminating between people and groups concerning their understanding of gene technology. It might not be the best indicator of knowledge, since there is a range of ways of defining and measuring knowledge, but we still find it useful. It is, however, important to stress that there is nothing to indicate that the relation between knowledge and attitudes is self-evident. Hisschemoller and Midden (1999) argue that the relationship depends on the evaluation of attributes connected to, in this case, a particular technology. Then, knowledge may influence the attributes that are part of forming the attitude, which in turn gives an indirect effect on the attitude-forming process.

¹⁴The variables were measured using the number of respondents reporting that they trust a specific institution doing a good job for society. This means that the number of variable values differs between the four groups of institutions depending on the number of specific actors included; experts range from 0 to 2, stakeholders from 0 to 4, NGOs from 0 to 3 and official bodies range from 0 to 2.

¹⁵All differences between countries are significant on the 5 percent level in a one-way ANOVA.

¹⁶Age is categorised in four groups; 15–24, 25–39, 40–54, 55 and older.

¹⁷However, education had to be excluded because of multi-colinearity with knowledge.

¹⁸Index of ten questions.

¹⁹The question was ‘Before today have you ever talked about modern biotechnology with anyone?’ and answered on a four-point scale from no never to yes frequently.

²⁰In the Eurobarometer survey a so-called split ballot was used when asking about different applications such as GM food and stem cell research, which means that only half of the sample answered these questions. As a result the sample is decreased by 50 percent of the total Eurobarometer sample.

²¹The chosen method order logit regression, does not measure explained variance since the purpose of this study was to confirm the significance of institutional trust rather than to test the strength of the relationship. Compared to the control variables, which have shown stable correlations in earlier studies, trust is the most stable and influential variable indicating reliable results.