The Trust Paradox: The Role of Context Effects in Stated Trust Judgements

Abstract

Why do people vote, read newspapers and buy mobile phones when they claim politicians, the press and the mobile phone industry can't be trusted? One account of this apparent paradox argues that people are aware of their ambivalence and only exhibit trust‐like behaviours with caution and scepticism. An alternative account suggests we underestimate the impact our implicit trust has on our behaviour when making explicit trust statements. The current research offers a third, not mutually exclusive, explanation. Specifically, it is argued that statements of trust are not simply reflections of some underlying level of trust but are systematically influenced by elicitation context effects. To date, most research has asked people to rate several targets simultaneously (joint evaluation (JE) context) rather than in isolation (separate evaluation (SE) context). It was predicted that this JE approach would encourage contrast effects between targets and lead to a greater spread of trust ratings than using a SE approach thus tending to amplify any stated‐revealed trust differences. Evidence was found by comparing trust ratings of various targets with respect to mobile phone technology risks across two studies one using a JE frame and one using a SE frame.

Keywords:

• Trust
• trust paradox
• critical trust
• context effects
• mobile phones

Acknowledgements

This research was supported by the UK's Health and Safety Executive, Contract No. 4101/R71.046 awarded to J. Richard Eiser (PI). Thanks also to Sabine Pahl, Jamie Wardman and two anonymous reviewers for helpful comments on previous versions of this article.

Notes

1. The terms are borrowed from Tversky and Griffin's (2000) analysis of personal well‐being judgments. For them, an endowment effect refers to an event's “direct contribution to one's happiness... positive experiences enrich our lives...hard times diminish our well‐being” (p. 709). However, “events also exercise an indirect contrast effect on the evaluation of subsequent experiences. A positive experience makes us happy, but it also renders similar experiences less exciting” (p. 709). In the current context, an endowment effect would reflect a general level of trust, being higher in one target than another, irrespective of context. Independent scientists, for instance, tend to be trusted more than industry representatives irrespective of context. A contrast effect would be that amount of variance in trust judgments attributable to the context such as the range of targets to be evaluated.

2. Hsee's focus on joint vs. separate evaluation (JE vs. SE) contexts was somewhat different (Hsee, 1996; Hsee et al., 1999). He was primarily interested in the possibility that preference reversals across SE and JE contexts could occur because some features of an object are harder to evaluate independently in an SE context (e.g., the number of entries in a dictionary) than other features (e.g., whether the dictionary has a damaged cover). Although the current research is thus only tangentially related to Hsee's, the JE vs. SE terminology neatly summarises the distinction between the contexts of current interest in terms of the number of targets considered.