Fisher's exact and Yates' correction for continuity

#1
i got a bit confused, could any one remine me please, when to use Fisher's exact and Yates' correction for continuity in 2x2 tale with one or more cells lessthan five?

are we using yates for n more than 20 0r 40 and fishers when n less than 20 or 40

completley lost

please direct me
 

Dragan

Super Moderator
#3
i got a bit confused, could any one remine me please, when to use Fisher's exact and Yates' correction for continuity in 2x2 tale with one or more cells lessthan five?

are we using yates for n more than 20 0r 40 and fishers when n less than 20 or 40

completley lost

please direct me
You're right, you are confused :). That is, the concern to be raised is when the expected frequencies (not the observed frequencies) are less than 5 in the context of Pearson's chi-square statistic test of independence for a 2x2 design.

Why not just use Fisher's exact test (forget about Yates' correction) - and I suspect this might solve your problem.