Statistics gurus... simple question

[FP]

I was reading an argument on reddit today and it got me wondering...

The issue was about how much caffeine would kill you.

Someone posted a link to a site that said the median amount of caffeine that would kill you is... let's say 15mg.

Someone else then said "so if I ingest 15mg of caffeine I have a 50/50 chance of dying."

That made sense to me. 50% of the population is killed by 15mg or more of the stuff.

But someone else replied that he is confusing the median and mean and that the probability of dying is not the same... leaving me completely confused.

So who was right?

 

[FleshLight]

http://en.wikipedia.org/wiki/L.D._50_%28album)

 

[GodlessAstronomer]

Median is correct, lethal dosages are quoted in LD50, the lethal dosage for 50% of the population. But saying you have a 50/50 chance of dieing is simplistic - it ignores the factors that go into that number, like body weight and health issues.

 

[JJChicken]

But someone else replied that he is confusing the median and mean and that the probability of dying is not the same... leaving me completely confused.

This guy is correct.

This is because the probability of dying is not constant to the amount of caffeine that would kill you, i.e. more caffeine you take, you have a higher chance of dying.

 

[JJChicken]

Median is correct, lethal dosages are quoted in LD50, the lethal dosage for 50% of the population. But saying you have a 50/50 chance of dieing is simplistic - it ignores the factors that go into that number, like body weight and health issues.

Didn't know about this convention, if thats the case, then I retract my answer :'(

 

[GodlessAstronomer]

http://en.wikipedia.org/wiki/L.D._50_%28album)

I used to love this album when I was a pot-smoking burn-out teenager, but I don't think it's what the OP was looking for