A random sample of 16 mid-sized cars tested for fuel consumption gave a mean of 26.4 kilometers per liter with a standard deviation of 2.3 kilometers per liter.

i) Assuming that the kilometers per liter given by all mid-sized cars have a normal distribution, find a 99% confidence interval for the population mean .

ii) Suppose the confidence interval obtained in (b)(i) is too wide. How can the width of this interval be reduced? Describe all possible alternatives. Which alternative is the best and why?

i have found the answer for (i) and that is:

Mu = 26.4

sigma = 2.3km per liter

N = 16

confidence interval = 99%

so substituting this infor in the confidence interval formula,

x bar +- z alfa/2(sigma/root of N)

26.4 - 2.58(2.3/root 16)

24.92

can anyone explain me how can i solve for(ii)....plizzz