Displaying 1 result from an estimated 1 matches for "sex_ratiob".
Did you mean:
sex_ratio
2011 Jul 21
0
gls yields much smaller std. errors with different base for contrasts
...04 0.5257303 2.9974788
Residual standard error: 21.90841
Degrees of freedom: 36 total; 30 residual
1> levels(tmp.data$sex_ratio)
[1] "3m:9f" "6m:0f" "6m:6f"
1> #Now change the contrasts so the base level is the all male sex
ratio (i.e., 6m:0f)
1> tmp.data$sex_ratiob<-tmp.data$sex_ratio
1> contrasts(tmp.data$sex_ratio) #original contrasts
6m:0f 6m:6f
3m:9f 0 0
6m:0f 1 0
6m:6f 0 1
1> #Set new contrasts for the copied sex_ratio column (i.e., "sex_ratiob")
1> contrasts(tmp.data$sex_ratiob)<-contr.treatment(n=...