Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Independent sample t-test right?

0 views
Skip to first unread message

cambodia

unread,
Dec 13, 2009, 10:37:26 PM12/13/09
to
i am trying to analyse data i collected on head measurements of storks
and somehow i have the feeling my results are just wrong (and i'm not
a pro on stats, just in case you didn't realize after reading this)...
(;
my hypothesis is that there is no difference between female and male
storks concerning those measurement i took from digital pictures (i.e.
the values are not real, but i calculated the ratios to make them
comparable).
now i've run an independent sample t-test on the data, with male=1 and
female=2 as grouping variable and all the separate ratios for each
stork as test variables.
the Sig.(2-tailed) i get for each ratio seems not right as they all
show a value between 0.2 up to 0.9......means there's no sig
difference at all? also the boundaries of the confidence interval show
values for Lower: around -0.01 and for Upper: around 0.01. so the
hypothesis value(0) falls within those intervals which shouldnt be the
case, right??
did i may violate some assumptions...do i have to use a nonparametric
test? or could i've done somethign wrong with the variable properties
or alike...??
thanks for any help or suggestions!
0 new messages