Study questions
- Write a summary of the results given in this case study
in about 100 -150 words describing the difference in weaning weight among
genotypes and the effects of the various covariates on weaning
weight.
- Fit the final statistical model again and save the
residuals. Produce the boxplots by genotype and age of dam, as done earlier in
this case study. How many outliers are now shown? What will you do about
them?
- Fit an additional cubic term (DC) for DAMAGE7 to the
model and examine its statistical significance. Based on what you find would you
include a cubic term or not?
- We have observed the partial confounding between
genotype and year. Run the model again without year and comment on the parameter
estimates and their standard errors that you now find for genotype. Calculate
the new value for the average standard error of difference between pairs of
genotype means and compare with the earlier value.
- Include BIRTHWT as a covariate in the model. What
effect does this have on the analysis? Why? Using Genstat to calculate an analysis of variance to compare
birth weights among genotypes may help to explain what has happened. Based on
this do you think it appropriate to include BIRTHWT in the model and, if so,
what interpretation do you put on the results?
|
|