the mean percent defective in the samples will be the same as the percent defective in the population, so it's 5%

the standard deviation is the square root of (n*p*q) where q=1-p

use the normal approximation to the binomial to determine the probability of getting 10% defective or more in a sample