- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
Hello all,
I have a query if I have same sample size within different years example 100 lines but replicates of them vary in years for an example one year has 6 replicates while other years have 3 replicates. In this case my data is balanced or unbalanced?
what method on SAS I should use to calculate ANOVA for such data?
Thanks
Accepted Solutions
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
You've probably got more than that going on, if you are interested in heritabilities. I assume that the four locations mentioned are all different? If so, you can ignore year as an effect, as it is completely confounded with location. If not, then you need to consider year-location as a fixed effect to remove from your heritability estimates. So, what are the other factors involved here? I have a hunch you are going to need one of the mixed model PROCs to get correctly calculated variance components you need to calculate heritablility.
SteveDenham
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
The data are unbalanced. Any of the linear modeling PROC's except PROC ANOVA could be used (GLM, MIXED, GENMOD, GLIMMIX, etc.) We would need more information on the design to make a good recommendation.
SteveDenham
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
Hello Steve,
Thanks for the reply. To be more precise I have observation A within three years. The number of observations are same i.e. 100 within all three years.
-First year has one location where observations were taken and replicates in that year are 6.
-Second year has one location but 3 replicates
-Third year has two locations and 3 replicates per location
Can you please help me if I want to check homogeneity of variance, ANOVA and heritability on SAS how can I perform it?
Thanks
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
You've probably got more than that going on, if you are interested in heritabilities. I assume that the four locations mentioned are all different? If so, you can ignore year as an effect, as it is completely confounded with location. If not, then you need to consider year-location as a fixed effect to remove from your heritability estimates. So, what are the other factors involved here? I have a hunch you are going to need one of the mixed model PROCs to get correctly calculated variance components you need to calculate heritablility.
SteveDenham
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
Thanks Steven