How do you interpret the area under the ROC curve (AUC)?

• A. The slope of the ROC curve at the chosen cutpoint
• B. Probability that a randomly selected diseased animal has a higher test value than a randomly selected non-diseased
• C. The difference between sensitivity and specificity
• D. Probability that a random animal tests positive

Answer: (B)

The key idea is how well the test separates diseased from non-diseased across all possible thresholds. The AUC represents the probability that a randomly chosen diseased animal will have a higher test value than a randomly chosen non-diseased animal. In other words, it’s a measure of the test’s ability to rank diseased above nondiseased regardless of where you set the cutpoint, and it corresponds to the Wilcoxon rank-sum concept.

An AUC of 0.5 means no discrimination (random ranking), while an AUC of 1.0 means perfect separation. It’s not about the slope at a specific cutpoint, nor the difference between sensitivity and specificity at one threshold, nor the chance of testing positive at a given cutpoint.