Plan a sampling experiment to partition the source of variability in auto gas mileage in your community between drivers, cars, and replicates.

(a) Describe how you would plan the study if drivers and cars were crossed factors.

(b) Describe how you would plan the study if drivers were nested within cars.

(c) Determine the number of degrees of freedom you would need for the error or replicate term in your model if you desire the width of the condence interval on the variance component for replicate to be 75% of 2.