管理员
- 积分
- 77135
- 大米
- 颗
- 鳄梨
- 个
- 水井
- 尺
- 蓝莓
- 颗
- 萝卜
- 根
- 小米
- 粒
- 学分
- 个
- 注册时间
- 2009-5-28
- 最后登录
- 1970-1-1
|
en 作为statistician 天天给lay people 解释statistics,我体会的一个问题是,even if all things can be very accurately measured, annotated and via a complex formula ,都归结成一个简单的概率了,人们还是无法正确理解之。
比如说,woman A, with your DNA like this, and your life style like this, plus any and all other variables, we estimate the probability of you developing breast cancer by the age 45 to be 75% percent.
What exactly does this mean? It means for a large number of women, say 1000, all with exactly the same DNA make up, same life style and same on almost all other dimensions, roughly 750 of them will go on to have breast cancer by age 45.
But for woman A, this 75% doesn't say for sure whether or not she will have breast cancer by 45.
Very commonly upon hearing this, someone will go on to *enlighten* me "but a probability is meaningless. Either you will have it or you won't. The probability is either 1 or 0."
it's still not easy for people to understand the difference between characteristic of a random variable (it's probability distribution) and one realization of it.
And that is in fact the truth about evidence based medicine, it doesn't tell you whether or not you will be cured by this particular drug, only that a certain percentage of people who take the drug will be cured.
Then there's the problem of choice and decision, after understanding the hard facts.
Say a couple finds out their unborn child has "90% chance" of having this rare and terrible genetic disease, should they choose to terminate the pregnancy?
If their new born is alive and suffers from brain damage and very unlikely to recover from it, should they remove the ventilator and let the baby die?
People instinctively believe in the benefit of having the freedom to choose, but some studies have shown that in certain situations more choice does not necessarily lead to more happiness or more confidence/satisfaction about their choice.
And there's the further problem of "bandwidth". How much time /energy should we spend on understanding the facts enough to make sound judgments?Is the time/energy expenditure justified by the benefits we stand to reap from making "slightly better judgments"? Or are we overwhelmed by the sea of information, the fear of inability to understand, the anxiety about drawing the right conclusion from our understanding, or the guilt about not understanding them....
We can call for aid from these rational computing machines. But we are not such machines. We are subject to fallacies, emotions, biases, and all those recently popular terms. Maximizing self perceived happiness on a moment to moment basis is a rather different task from making the best scientific data-based judgments..... |
|