Trust is the foundation for everything. With trust, you have creativity, productivity, and financial results. With trust, you have strong marriages and healthy relationships with your kids, parents, and friends. But how do you measure trust? Trust is a relatively intangible, abstract construct that’s hard to measure. You can’t just send a questionnaire and ask people, “what is the level of trust in your organization?” Trust is relative, and therefore no absolute answer would be accurate. Yes, if trust is the foundation for everything, you must measure it somehow. In this article, I will provide a philosophy of measuring trust.
Why is Trust Hard to Measure?
Is there such a thing as the overall level of trust in a team, a company, or any other group of people? I believe there is, but it’s not something that can be measured overall. Since trust is personal and asymmetrical, there are 6 relationships between 3 people. Each has a level of trust in the other two, independently of any other trust level. Why is that important? Consider this: you are in a small team with 3 other members. You have a high level of trust in each of them. How comfortable are you arguing passionately with them? How about being vulnerable with them? Asking stupid questions? Suggesting stupid ideas? Asking tough questions? Giving them direct, unfiltered, no-BS feedback? You probably feel comfortable doing all those things. Now, imagine that a fifth member is added to the team. Whether it’s because you don’t know that member very well, or for any other reason—you don’t trust that member very much. How comfortable are you doing all or any of those things? Your comfort level probably took a nosedive simply by adding this new team member.
Therefore, I would claim that the level of trust in a team is measured as the average of the lowest levels of trust that each member has in any of the other members.
Trust is relative. It is contextual (2nd Law of Trust) and personal (3rd Law of Trust), and therefore measuring trust in the team must start with measuring every single relationship on that team.
But, how much trust do you need on the team? It depends (no pun intended) on each member’s dependency on the other team members. If team member A doesn’t depend on the work of team member C, it doesn’t matter whether A trusts C. But if A depends on what C does in the team yet doesn’t trust C, then we have a problem. So, measuring trust must be done in the context of the dependency between members.
And, given that the level of trust is personal, the level of trust required is for every team member to feel safe. If the level of trust you have in another team member compensates for the subjective fear of an objective risk you are facing, you feel safe. If the level of trust is not enough to compensate for that fear—you feel danger.
How to Measure Trust: The Black Box Approach
Since trust and trustworthiness are hard to measure, we must resort to the “black box” measurement technique. To explain that technique, I will use an electric motor as an example. Measuring the magnetic field inside the motor is hard. However, a motor has inputs and outputs. We can consider the motor a black box (you don’t know what’s inside). The input to that box/motor is electricity (voltage, current, power, frequency, etc.). The output is mechanical (force, torque, speed, RPM, etc.). Those inputs and outputs are much easier to measure than the magnetic fields inside the motor. Measuring the inputs and outputs would give a strong and accurate indication of the magnetic fields inside the motor.
To apply the black-box method to trust, I’ll start by reviewing the “trust chain.” It starts with the 6 components of trustworthiness (more in the next section). Their existence creates trustworthiness. Combined with trustability (the other person’s willingness to trust anyone), we get trust. Trust’s existence (or lack thereof) would drive autonomy, accountability, the willingness to hold constructive disagreement, vulnerability, etc. The inputs (the 6 components) are relatively easier to measure. The outputs are the symptoms of trust’s existence (or lack thereof) and are relatively easier to measure.
So, if we measure high levels of inputs, we can infer that trustworthiness and trust both exist, which will be further supported by the existence of the outputs, which serve as trailing indicators.
Inputs and Outputs of Trust
However, we must keep in mind that the measurements of the inputs and outputs are subjective (personal) and contextual. They should be measured from the perspective of different individuals, the way they see them, and in the context of what you want to trust the other person with. Take the outputs, for example. If you wish to measure whether leaders trust their followers, you will use outputs such as autonomy vs. bureaucracy. Does the leader give their employees autonomy to do their jobs, or do they micromanage the employees? If you wish to measure whether an employee trusts their leader, you would measure outputs such as accountability vs. CYA. In general, you could always measure outputs such as the willingness to be vulnerable, give feedback, etc.
Measuring Trust is More Qualitative Than Quantitative
We love multiple-choice surveys. They are easy to handle, and they are scalable. You can send a survey to hundreds, thousands, or more participants and easily deploy statistical procedures on the results file. But given the nature of trust, being relative, personal, and contextual, a simple survey might not do the trick. Sure, certain components generally indicate trust, but those are few and far between.
Instead, I would suggest that measuring trust be done on a qualitative-exploratory basis. Instead of asking multiple-choice questions, you should ask person A what makes them consider a person in C’s position competent (as an example, for one of the 6 components). Ask for the most important things. The odds are that the precision level of the answer would be something you would not have come up with yourself. Then, once you know the most important factors that person A considers to show a person’s competence in C’s role, ask them to rate person C in those factors from “the best there is” to “the worst there is.”
This may require interviews, limiting the scope of your group trust assessment but could also be done through a questionnaire that has a space to write the answers instead of having multiple-choice answers. Then, either through people reviewing or coding the answers or using artificial intelligence tools, calculate the level of trust.
Sounds hard? That’s the nature of measuring trust.
Dr. Yoram Solomon is a trust expert, author of The Book of Trust, host of The Trust Show podcast, a two-time TEDx speaker, and facilitator of the Trust Habits workshop that helps building trust in organizations.