In which of the following instances would the t-distribution be used to model the sampling distribution of a single sample statistic instead of the z-distribution?
(A) Constructing a confidence interval for the population proportion; the population proportion, p, is unknown.
(B) Conducting a matched-pairs hypothesis test for the population mean difference; the population standard deviation of the difference, sd, is known.
(C) Constructing a confidence interval for the difference between two population means; the population standard deviations, s1 and s2, are unknown.
(D) Conducting a hypothesis test for the difference between two population proportions; the population proportions p1 and p2, are unknown.
(E) Constructing a confidence interval for a population mean; the population standard deviation, s, is known.
Answer: C