A positive signal: what base rates change in decision-making
In pharmaceuticals, life sciences or Rhine logistics, a test result or an alert can trigger a substantial commitment of resources. The managerial question is specific: before committing resources on the strength of a signal, how should the underlying frequency of what it indicates be taken into account?
What the literature establishes
Daniel Kahneman and Amos Tversky (1973, Psychological Review) showed that predictive judgements can rely on how closely a case resembles a category rather than on the information needed for a probabilistic estimate. In their experiments, participants sometimes gave too little weight to base rates: the underlying frequencies of the categories under consideration. For a manager, a compelling case therefore does not remove the need to ask which reference population it belongs to.
The denominator a signal can obscure
Gerd Gigerenzer and Ulrich Hoffrage (1995, Psychological Review) showed that presenting information as natural frequencies can facilitate Bayesian reasoning without prior instruction in the rule. The principle is to describe, within a single reference population, how many cases exhibit the phenomenon of interest and how many generate a positive signal, with or without that phenomenon. This representation makes an essential distinction easier to grasp: detecting a phenomenon when it exists is not the same as establishing its presence when a signal appears.
The premature conclusion
One might conclude that replacing percentages with counts is enough. That would overlook the fact that natural frequencies preserve the structure of subgroups within a common population: raw counts drawn from different populations do not offer the same benefit. Nor does a better-understood estimate determine the action by itself: the costs of false alarms and missed cases still require a trade-off. (our executive and employee training programmes)
What the available evidence does not guarantee
These studies concern experimental judgement tasks, not the effectiveness of a decision-making system in a Basel company. In practice, the base rate may be poorly documented, vary across sites or come from an already selected sample. An international R&D team therefore needs to make the reference population and data uncertainty explicit before turning a signal into a recommendation.
A practical check in Bâle
As part of SHR's “Decision-making” programme, a Basel team in pharmaceuticals, life sciences or Rhine logistics could review resource commitments based on a test or an alert over a period defined in advance. For each case, it would record the reference population, the base rate and its source, and the available information on false positives and missed cases, explicitly identifying unknowns. The measure would be the proportion of cases in which these elements are documented before the decision, verifiable in dated records. This indicator measures preparation discipline, not a demonstrated improvement in outcomes. To go further: explore the Decision-making training in Basel, or browse our executive and employee training programmes in Switzerland.
In pictures: Decision-making in Basel
- decision-making
- Basel
- research
- training basel