hi
i would like to the non parametric equivalent of two way
analysis of variance with var 2 as the dependent variable
and group as the treatment variable and block as the block
variable - does anyone know of a program that will do
friedman's equivalent of two way anova - i have looked at
minitab and the block sizes are equal to 1 - as you can see
fro the below from the aggregated data there are more than
1 entry for the same block
if anyone knows of a method of carrying this pout in either
bmdp o9r genstat please could they also let me habe an
example of the code necessary to carry out the procedure -
i am trying to ascertain
a if there is adifference in median (grouped because of
ties) between the groups
b if there is a factor (called block) which adjusts the
grouped median value of the groups i.e. is there a
difference in the grouped medians of var2 after adjusting
for a second factor (block)
any help on this would be much appreciated
block var2 group count
1 42 3.00 28
1 42 5.00 15
1 42 6.00 17
2 42 3.00 9
2 42 5.00 3
2 42 6.00 24
3 42 3.00 4
3 42 5.00 2
3 42 6.00 5
4 42 3.00 3
1 43 3.00 3
1 43 6.00 1
3 43 3.00 1
3 43 6.00 2
regards
ann
----------------------
ann e fitzmaurice
medical statistician
dugald baird centre
university of aberdeen
[log in to unmask]
tel 01224 681818 extension 53876
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|