Re: Significance testing of Dependent proportions

Hari wrote: 
Hi Bruce,

Let me take the data from the link you provided me long ago when I
wanted to do significance testing of Independent proportions (
http://www.gseis.ucla.edu/courses/ed230a2/notes3/portion.html )

Group       n        Frequency              p
Group1      261        127                   0.487
Group2      160          65                    0.400

My doubt is, if I want to do significance testing on the above data and
if I know that Group1 and Group2 are NOT independent then how do I go
about it?

HOW are they not independent? Matched pairs of subjects? Two observations per person?

--
Bruce Weaver
bweaver@xxxxxxxxxxxx
www.angelfire.com/wv/bwhomedir
.