s******n 发帖数: 95 | 1 I have 4 independent groups, one is the control group and the other three
are treatment groups.
and I have 31 study days, so I have a couple of data for every particular
day in every group.
so it's something like this:
day 1 day 2.....................................day
31
control group a e ......................
b f ...........
c g .........
d h . | s******n 发帖数: 95 | | l******r 发帖数: 18699 | 3 if you test it for each particular day,which means you drop the day effect,
then why use a repeated measurement model?
I guess in most of the situations, people tend to consider day effect,
otherwise, some useful information will be lost. in this case, a repeated
measurement anova is a good choice
day
【在 s******n 的大作中提到】 : I have 4 independent groups, one is the control group and the other three : are treatment groups. : and I have 31 study days, so I have a couple of data for every particular : day in every group. : so it's something like this: : day 1 day 2.....................................day : 31 : control group a e ...................... : b f ........... : c g .........
| b*****n 发帖数: 685 | 4 有missing data你的ANOVA就不好用了。这个是longitudinal data的hypothesis test
问题,有点麻烦,不过parametric和nonpara的方法应该都有,查查paper了,比如
Jianqing Fan的。 | D******n 发帖数: 2836 | 5 repeated measurement and then do contrast at each specific day levels with
multiple comparison adjustment.
but as some1 pointed out, why are u so interested in each particular day
level? it is usually not the question to ask.
day
【在 s******n 的大作中提到】 : I have 4 independent groups, one is the control group and the other three : are treatment groups. : and I have 31 study days, so I have a couple of data for every particular : day in every group. : so it's something like this: : day 1 day 2.....................................day : 31 : control group a e ...................... : b f ........... : c g .........
|
|