7.3 Schall (1991) considers data from an experiment to measure the mortality of cancer cells under radiation. Four hundred cells were placed on a dish, and three dishes were irradiated at a time, or occasion. The zero-dose data provided here were obtained from OzDASL and are found in cell__survive.xls.

(a) Consider occasion of irradiation as a fixed effect and assume that the three dishes irradiated at a given time represent a random sample of possible dishes of cells that could be irradiated at a given time. Fit an appropriate logistic regression model to determine if there is a significant difference among the occasions.

(b) Now assume that both occasion and dish are random effects and re-fit your logistic model to determine if there is a difference between the occasions.

(c) How does the interpretation of your hypothesis test in part

(b) differ from the interpretation of the hypothesis test in part (a)?

(d) Provide an estimate of the overall proportion of cells that survive using the model in part

(a) as well as the model in part (b).