Despite decades of research into trust measuring individual trust remains unsatisfying due to problematic survey questions that are used to measure social trust. There are two main methods used to measure trust, surveys and behavior observation. Surveys ask for people’s judgments about trust, while behavior observation looks at behaviors based on trust. The trust game is an example of the latter, where trust is measured by the amount of money sent from the trustor to the trustee and how both parties behave.
TrustTalk podcast host Severin de Wit talked about that with Paul C. Bauer, a Fellow at the Mannheim Centre for European Social Research for this 62nd episode of the TrustTalk podcast. He wrote his Ph.D. (“Three Essays on the Concept of Trust and its Foundations“) on trust and is currently involved in innovative, audio-driven trust measurement.
Trust reports by major consultancies and non-alignment
He suggests that non-alignment between the many surveys by organizations like Edelman, Pew Research Center, and major consultancies like Deloitte, PwC, EY and KPMG may be due to different survey questions, different time points for data collection, and different samples. The OECD Guidelines on measuring trust are helpful but could be updated with more recent knowledge. He believes that even though trust is emotional and subjective, it should still be quantified, and mentions his research on audio responses to measure trust. He also points out that definitions of trust are often detached from measurements of trust and discusses his work on finding a better fit between the definition and measurement of trust.
(…) we simply learned over time that the questions we use are to some extent problematic. So, for instance, the most popular survey question that we use to measure social trust goes along the lines: “Do you think that most people can be trusted or that you can’t be careful enough in dealing with people?” And this question contains concepts such as most people, And by now we know that people might interpret this question differently. This is a measurement problem, and we find such measurement problems across different trust survey questions.
Behavioral Trust
(…) So, for instance, I could do an interview with you because I trust you, and in sociology and political science, the most common method to measure trust is to ask people for their judgments in surveys. So we could ask them: “Do you trust your family? Do you trust your neighbours?” and the infamous generalized social trust question that I just mentioned is, for instance, “do you think that most people can be trusted or that you can’t be too careful in dealing with people?”. That’s one way, basically directly asking for these judgments. And besides that, we can also observe the behaviour and then conclude that certain behaviours we observe should be based on a high level of trust. So, for instance, I could observe a friend lending money to another person, and then I would assume that she trusts this other person because I’ve observed this behaviour and there’s actually a large literature that attempts to measure what trust researchers call “behaviourally exhibited trust”, in other words, behaviour that is based on trust.
Transcript of Interview Paul Bauer
YouTube -interview with English subtitles
Contact Paul C. Bauer on Twitter
Want to contact Paul Bauer or follow him on Twitter (@p_c_bauer)
Support TrustTalk, enable us to make more podcasts
Do you like our TrustTalk show? Help us by donating whatever you like to give us. By doing so, you help us to continue to make more episodes in 2023. That is possible on https://trusttalk.co/donate. Thank you very much for your support! We greatly appreciate it.