How likely is likely

Anything that is just likely is likely wrong

The quote comes from the French philosopher, mathematician and natural scientist René Descartes. In general, we understand probability to be a classification of statements and judgments according to the degree of certainty (). This judgment about likelihood is a construct, as reality is permanently created by our highly subjective perceptions. Our perception of risk, in turn, depends on what our senses condense into an overall picture. Our knowledge, emotions, morals, fashions, judgments and opinions determine this construct. What one person perceives as doesn’t have to be for the other.

Furthermore, risk perception is based on hypotheses. As a result, different assumptions and theories are often made for the same risks. The discussion about the risks of genetic engineering is an example of the subjectivity of (social) risk perception. On the one hand, resistance can be observed as a protest against being overwhelmed by innovation processes and based on fundamental ethical objections. On the other hand, the opportunities in plant breeding, animal breeding, the food industry and medicine are "seized". Perception, in turn, is determined by a context, i. H. the consideration of the space and time perspective. There is no unrelated perception. Reality therefore remains an illusion, since there can be no "wrong" or "right" in the world of perception.

In short: our human brain is incapable of dealing with probabilities - and yet we work with probability statements in the practice of risk management. We regularly torture our risk officers with statements about probabilities and present the board with risk maps with statements about the probability of occurrence. We know that we lack an intuitive understanding of probabilities. Or can you explain the difference between the 62 and 69 percent probability of occurrence? In science, this is referred to as "neglect of probability". In practice, this leads to decision-making errors and regularly also to crises.

They do not believe me? A simple example is provided by the birthday paradox: There are 23 people on a football pitch (twice 11 players and a referee). The probability that at least two people will have their birthday on the same day is greater than 50 percent.

When it comes to investments, we mainly compare returns. The different risks are often ignored. The most recent financial and currency crisis has provided us with evidence that we don't have a natural sense of risk - and that we prefer to ignore it right away. Among other things, the anchoring effect from cognitive psychology provides evidence. In particular, environmental information has an influence on the (-) assessment even if it is actually irrelevant for the decision. In short: we orientate ourselves on an arbitrary "anchor": it becomes a construct of our perceptions.

But what do I learn from this for corporate practice? Stop tormenting your colleagues with questions about probabilities. For example, work with methods of the. There you are not asking about a probability, but about a range of effects. The underlying triangular distribution (or Simpson distribution) only requires information on the minimum value (worst case), the maximum value (best case) and the most probable value (realistic case). The density function looks like a triangle. The y-axis shows the respective probability for a value x∈ [a, b].

Conclusion: We should be careful not to cover up ignorance with only supposedly exact probability statements.



[Image source: iStockPhoto]