Quote:
Originally Posted by Melkerson
That was just one example to try to illustrate a point that you were for some reason not getting.
I think we also touched on the invincible, dumb 18 year olds. You could do all sorts of unnecessary tests on many of them and they would give no ****s, regardless of results. Even if the results were very bad and needed serious attention, they would likely still give no ****s. Stupidity is protective in this case.
We can construct all sorts of examples. We are only limited by our imaginations.
The examples would need to fit the criteria for probability and realism.
We have, as requirements, that risks are assessed PRIOR to knowing the results. Tbis includes NOT knowing how a teen will react, even if we have some indications of how they MIGHT react from doing a complete psychological assessment, interviews with their friends and family, etc.
If 10% of the imaginary teens you mention do x, then there is a risk of x. If one percent of them do x, there is a risk of x. If one tenth of a percent of them do x, there is a risk of x. I could go on, but I hope that you have recognized the pattern.
We do not, under any circumstances, get to be results oriented for a data point when determining risk, since, again by definition, the results MUST be unknown when determining (or guestimating) probabilities or it simply is not a probability problem. We simply cannot look at the result and then go backwards to say "see? No risk here." To do so would break every law of probability and statistics.