Study questions
-
Choose model b) in Question 4. Write a short report including a description of the statistical analysis (with the statistical model expressed algebraically) and a summary of the results.
- Explain what is meant by fixed and random effects. Give an example of a split-plot design. Describe the random effects that feature in its analysis. Why can these random effects be estimated by a conventional least squares analysis of variance so that REML is not necessary in this case?
- Yields from a number of plots are missing in a split-plot design. Describe two alternative ways of analysing the data and say when you think one method is preferable to the other.
- What do you understand by 'heritability'? The formula for calculating an estimate of heritability from a number of offspring nested within sires is
4ss2 / (ss2 +
se2), where ss2 is the variance component among sires and
se2 is the error or residual component. Calculate the heritability estimate from the REML output and also its standard error. A heritability estimate lies between 0 and 1. Comment on the result you get.
|
|