Re: group-based judgement

From: Louis Newstrom (louisnews@comcast.net)
Date: Thu May 30 2002 - 17:50:32 MDT


I don't know who originally wrote:
> > >>> You collect two kinds of mushrooms, the
> > >>> fleebs and the kloongs. 10% of the fleebs will give you
> > >>> diarrhea, whereas 20% of the kloongs will. You know that
> > >>> most will not harm you at all, but you know that devouring
> > >>> a kloong is twice as dangerous as eating a fleeb.

> > > But lets not get all mathematical. {8^D

Actually I am going to get very mathematical.

I am making two assumptions that were not specified:
1) the two kinds of mushrooms are equally common.
2) it only takes one bad mushroom to make you sick

Original situation:
Say you have some mushrooms, and do not discriminate.
15% mushrooms are bad.
85% mushrooms are good.
only 20% chance you'll get ten good ones in a single meal (85% ^ 10).

So originally you had:
- 20% good meals
- 80% meals that make you sick

The proposed solution (that I am going to reject later) was to throw out all
kloongs because they're "twice as bad".

New situation:
10% fleebs are bad.
90% fleebs are good.
35% chance you'll get ten good fleebs in a single meal (90% ^ 10).

So what you end up with:
- 17.5% good meals (all fleeb)
- 32.5% meals that make you sick (all fleeb)
- 50.0% meals in the garbage (all kloong)

You DO reduce meals that make you sick, but mostly by skipping meals. (With
that approach you could reduce all bad meals by not eating!)

This approach makes you go hungry 50% of the time, and gives you 2.5% LESS
good meals. It's hardly an improvement.



This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 09:14:31 MST