Computing the sample variance by dividing SS by n or dividing SS by (n – 1) has been debated. Which of the following explains the rationale for choosing to compute the sample variance as SS divided by (n – 1)?

a. It is more important that a sample variance varies minimally from the population variance, than that it is an unbiased estimator of the population variance.

b. It is more important that a sample variance is an unbiased estimator of the population variance, than that it varies minimally from the population variance.

c. There is no rationale needed; the sample variance is always computed as SS divided by n, not SS divided by (n – 1).