OTOH, climate science is a great example of scientists trying their damnedest to be conservative and euphemistic in their statements, e.g. "Future temperature in southwest Asia projected to exceed a threshold for human adaptability"(*), resulting in doing too-little/too-late and continuously having to revise upwards because things "surprisingly" always seem to proceed faster and be worse than initially reported.Ego wrote:When they start feeding the fear cycle we are all doomed.
(*) Military planners would phrase that somewhat differently. (Count how many times nuclear war is mentioned in this report.)
Both are examples of in/advertently trying to skew the risk-response. The best(**) strategy is to report expectation values (in the mathematical sense) rather than lower or upper p-levels. That is, for accurate policy responses, scientists should be reporting at the 50% level, e.g. we're predicting a 50% that the correct value is higher than X and also 50% that it's lower than X. Whereas the conservative approach is to say, there's less than 5% chance that it will be worse than X.
(**) Of course "best" is a matter of perspective. Best as in best for what/whom?
The problem is that the public and policy planners (most of whom don't have a scientific background) don't pay attention to the p-levels which gives much more weight to the wording and the timing of the wording than scientists, who intuitively look at the p-levels, would do.