Comparing risks: what the number of cases does not tell us
Comparing transport incidents, quality deviations or experimental outcomes requires relating events to the opportunities for observing them. The management question is specific: how can we prevent a more striking case count from overshadowing the frequency relevant to the decision?
What the literature establishes
Lee A. Kirkpatrick and Seymour Epstein (1992, Journal of Personality and Social Psychology) studied choices between containers holding different proportions of winning tokens. Their experiments show that subjective assessments of the chance of winning can depend on how the ratio of winning tokens to all tokens is presented. Choices therefore do not always reflect mathematical probability alone.
Understanding the ratio is not always enough
Veronika Denes-Raj and Seymour Epstein (1994, Journal of Personality and Social Psychology) observed that participants could prefer a set containing more winning tokens but a lower proportion of winners. Some thus chose against their own assessment of the mathematically preferable option. The proposed mechanism is a conflict between the concrete appeal of the number of possible wins and an analytical assessment of the ratio.
The premature conclusion
One might conclude that absolute counts should be removed from decision dashboards. That would confuse two questions: the probability of an event and its total volume. In pharmaceuticals and logistics alike, a large volume of incidents can justify action even when the rate is low; counts and frequencies should therefore both be presented, without treating them as interchangeable. (our executive and employee training programmes)
What these experiments do not demonstrate
These studies concern experimental probability-choice tasks, not decisions by managers in Basel. They do not establish that a common denominator alone improves industrial decisions. Apparently comparable rates can also conceal differences in severity, exposure duration or the definition of an incident.
A practical check in Bâle
Within SHR's “Decision-making” programme, an exercise to test in Basel is to revisit a risk comparison from pharmaceuticals, life sciences, Rhine logistics or an international R&D team. For each option, present the observed count, the volume exposed and the rate expressed in a common unit, retaining the period, event definition and severity. Collect a recommendation before and after this reformatting, then calculate the proportion of recommendations changed with a justification explicitly referring to the denominator. This measures sensitivity to presentation format, without proving that the revised decisions are better. To go further: explore the Decision-making training in Basel in the canton of Basel-Stadt, or browse our executive and employee training programmes in Switzerland.
In pictures: Decision-making in Basel
- decision-making
- Basel
- research
- training basel