Power analysis for ROC curve analysis
When you want to evaluate the diagnostic value of a numerical measurement, you can calculate the minimum sample size required in each group for the expected AUC (Area Under the Curve) value to be statistically significant with this menu.
Required information
α error: shows the significance level of the p value. Choose between 0.05 and 0.01.
Strength (1-β): Select between 80% and 90% for targeted strength
AUC value: Enter an AUC value between 0 and 1.
Sample calculation: To calculate the minimum number of participants required to find a significant difference at the 0.05 level and with 80% power for an AUC value of 0.80, the information should be entered into the application as follows.
After checking Calculate, you can reach the minimum number of patients required. This field can be copied.
Reference: Obuchowski, Nancy A., Michael L. Lieber, and Frank H. Wians Jr. "ROC curves in clinical chemistry: uses, misuses, and possible solutions." Clinical chemistry 50.7 (2004): 1118-1125.