A health researcher samples patients from two clinics to compare their average wait times. What statistical test should be used to compare the mean wait times?

A health researcher samples patients from two clinics to compare their average wait times. What statistical test should be used to compare the mean wait times?

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

Answer: (A)

When you want to compare the average value of a continuous outcome across two groups that are independent, you use an independent-samples (two-sample) t-test. Here, the wait time is a continuous measure and the two clinics provide separate samples of patients, so the groups are not paired. The test determines whether the observed difference in mean waits is large enough to conclude a real difference in the populations, taking into account the variability and size of each sample. If variances are unknown, you can use the Welch version, which is still a two-sample t-test. The other options don’t fit: a one-sample z-test compares a sample mean to a known population mean with a known standard deviation; a paired t-test is for dependent or matched measurements on the same subjects; a 2-proportion z-test compares proportions, not means.

When you want to compare the average value of a continuous outcome across two groups that are independent, you use an independent-samples (two-sample) t-test. Here, the wait time is a continuous measure and the two clinics provide separate samples of patients, so the groups are not paired. The test determines whether the observed difference in mean waits is large enough to conclude a real difference in the populations, taking into account the variability and size of each sample. If variances are unknown, you can use the Welch version, which is still a two-sample t-test. The other options don’t fit: a one-sample z-test compares a sample mean to a known population mean with a known standard deviation; a paired t-test is for dependent or matched measurements on the same subjects; a 2-proportion z-test compares proportions, not means.