0
2

[–] Hydrocephalus 0 points 2 points (+2|-0) ago 

You mean you can fuck with variables until you get the result you want? Don't tell scientists! They might use it to help churn out publication after publication to help secure tenure...

0
1

[–] feral-toes [S] 0 points 1 points (+1|-0) ago 

Scientists are obsessed with Randomised Controlled Trials. Why? What are they so afraid of?

Why not just gather data from real life and analyse it. This is called epidemiology.

The parable illustrates how confusing and unhelpful the laws of arithmetic are, and how they brutally sabotage the efforts of epidemiologists to make valid deductions from non-experimental data.

0
0

[–] ShinyVoater 0 points 0 points (+0|-0) ago 

The ideal study has a group on the treatment, a group on a placebo and neither doctors nor patients that know who's in what group. Additionally, the patients should represent as diverse a population as possible in relatively equal amounts between control and treatment groups.

What this parable illustrates is the need for good analysis; in each case of the paradox, it comes down to looking at one variable at a time. If there really was such a wide array of good and bad outcomes, you'd make a chart of the success stories and start figuring out all the attributes they shared(and do the same for the failures, of course). Correlations drawn, you then design prescription guidelines that allow doctors to determine if there's any benefit to be had.