What does the ROC curve plot?

What does the ROC curve plot?

• A. Sensitivity versus 1 minus Specificity
• B. Specificity versus Precision
• C. Accuracy versus Error Rate
• D. PPV versus NPV

Answer: (A)

The ROC curve shows how well a diagnostic test can discriminate between positives and negatives as you change the decision threshold. It plots sensitivity (the true positive rate) on the vertical axis against 1 minus specificity (the false positive rate) on the horizontal axis. By varying the threshold, you get different pairs of TPR and FPR, which trace out the curve. A curve that rises toward the top-left corner indicates the test does a good job at identifying positives with few false alarms; the area under the curve (AUC) summarizes overall discriminative ability. Other options don’t fit because they pair metrics that aren’t arranged as a TPR versus FPR trade-off across thresholds. Specificity versus Precision, or PPV versus NPV, or Accuracy versus Error Rate describe different relationships and aren’t the standard ROC plot.

The ROC curve shows how well a diagnostic test can discriminate between positives and negatives as you change the decision threshold. It plots sensitivity (the true positive rate) on the vertical axis against 1 minus specificity (the false positive rate) on the horizontal axis. By varying the threshold, you get different pairs of TPR and FPR, which trace out the curve. A curve that rises toward the top-left corner indicates the test does a good job at identifying positives with few false alarms; the area under the curve (AUC) summarizes overall discriminative ability.

Other options don’t fit because they pair metrics that aren’t arranged as a TPR versus FPR trade-off across thresholds. Specificity versus Precision, or PPV versus NPV, or Accuracy versus Error Rate describe different relationships and aren’t the standard ROC plot.