basic proc mixed coding question

[deleted_user]

Hello all,
I'm trying to code a 3-way ANOVA model with a random effect using proc mixed, but I am unsure that I am specifying my model correctly. We have a series of environmental data collected from 3 types of sites (ref, deg, unk), each site is stratified into 2 reaches (near and far), and the data were collected over the 3 years. Multiple, distinct stations that represent one of the three types of sites throughout the estuary were selected randomly, and at each station 9 random replicate measures of the environmental data were collected. We are interested in detecting differences in the environmental data between ref, deg, and unk sites, as well as reach, year, and interactions of the 3. I want to make use of the 9 replicates at each station so I'm using the following model in proc mixed:

class = site reach year station
environmental data = site|reach|year
random site/subject=station

Is this the appropriate approach to take?

Thanks for any help.
-david

This is what happens when you don't have a proper statistician at a given institute.....

[SteveDenham]

David,

That approach looks pretty good, but there might be an even better approach. Is there any reason to believe that observations within a station and site are correlated from year to year? If so, you may want to expand the model to something like:

class = site reach year station
model environmental data = site|reach|year
random site/subject=station
repeated year/subject=station*site type=un

Good luck.

Steve Denham