American Board of Surgery Qualifying Exam (ABS QE) Practice Test

What is the correct test for non-parametric, paired quantitative data?

1. McNemar

2. Wilcoxon rank sum

3. Mann-Whitney

4. Unpaired t test

The correct answer is: McNemar

The McNemar test is designed specifically for analyzing paired nominal data, particularly in situations where you are interested in changes in responses or outcomes across two related groups. It is widely used in before-and-after studies or matched case-control studies, where the same subjects are evaluated under two different conditions or at two different times. For non-parametric tests, when you are dealing with paired quantitative data, the appropriate choice would actually be the Wilcoxon signed-rank test, which assesses differences between two related samples or repeated measurements on a single sample. This test is particularly useful when the data do not meet the parametric assumptions required for a t-test. The other options you provided do not suit non-parametric paired quantitative data. The Wilcoxon rank sum test and Mann-Whitney U test are useful for comparing two independent groups, rather than paired or matched data. An unpaired t-test, on the other hand, is specifically designed for normally distributed data with two independent groups, which is not applicable in this context of paired data. Therefore, while the McNemar test may seem to be an answer for non-parametric data, it is primarily suited for categorical data, not paired quantitative data. The correct test for non-parametric, paired quantitative