Which test would be used to examine the mean difference in a paired study?

Which test would be used to examine the mean difference in a paired study?

• A. 2-sample t-test
• B. 1-sample t-test
• C. Matched pairs t-test
• D. 1-proportion z-test

Answer: (C)

When data come from paired observations, you test whether there is a real average difference between the two conditions by focusing on the differences within each pair. Compute the difference for every pair, then test if the average of those differences is zero. This is done with a matched-pairs t-test. It leverages the pairing to reduce variability from individuals and increases power compared to treating the data as two independent samples. The alternative tests are not suited here: comparing two independent groups ignores the pairing; testing the mean of a single sample against a known value doesn’t address the difference between two conditions; and a z-test for proportions applies to categorical data, not means.

When data come from paired observations, you test whether there is a real average difference between the two conditions by focusing on the differences within each pair. Compute the difference for every pair, then test if the average of those differences is zero. This is done with a matched-pairs t-test. It leverages the pairing to reduce variability from individuals and increases power compared to treating the data as two independent samples.

The alternative tests are not suited here: comparing two independent groups ignores the pairing; testing the mean of a single sample against a known value doesn’t address the difference between two conditions; and a z-test for proportions applies to categorical data, not means.