TOPIC:

football for me that night

Weds 30th I've a friend and her boyfriend home briefly from Australia to drink with, but considering I'd been planning on recommending O'Neills to them anyway (she wants to bring the fella to a proper Irish pub, and others in the party have objected quite strongly to the B&C for some reason) that might work out grand <!-- s;) --><img src="{SMILIES_PATH}/icon_wink.gif" alt="" title="Wink" /><!-- s;) -->

Ok so when are we doing this?

Other then mocking people who have tastebuds of steel I do not think we will have enough people to get any useful data out of this. We are going to end up with a table something like
Person Guiness Beamish Murphys O'Haras
A 1 0 0 1
b 1 1 1 1
c 0 0 1 1
d 0 0 0 0

I think the expected numbers of rights if people guessed randomly would be

Guess all 4 right: 1/24
Guess 3 right: 0
Guess 2 right: 6/24 = 1/4
Guess 1 right: 8/24 = 1/3
Guess 0 right: 9/24 = 3/8

But we can work the statistics out afterward anyways.

So what's the minimum attendance for decent numbers?

I think to test the "people cannot tell the difference between stouts" hypothesis you need to do a chi-squared test on the results.
Say we put >=2 right in you can tell the difference and <2 in you cannot.
The results look something like this
#correct
_______ >=2 <2
#People 10 10

you need to have at least 5 in each of the #correct columns. Means you need at least 18 people.

Someone who has used or studied statistics should be able to correct me though.

&amp;quot;a_friend_in_mead&amp;quot;:hj9h7fb9 wrote: Means you need at least 18 people.[/quote:hj9h7fb9]
Right. I think we'll be lucky to get six. Will six do?