17.10 ROC Curve (Pro Only)

Introduction

The Receiver Operating Characteristic (ROC) Curve is used to represent the trade-off between the false-positive and true positive rates for every possible cutoff value.

By tradition, the false positive rate (1-Specificity) on the X axis and true positive rate (Sensitivity) on the Y axis are shown in the plot.

ROCCurve.png

Interpreting Results

ROC curves are used to consider whether a diagnostic test is good or bad.

We can judge the ROC curve from two criteria:

  • Shape
    If the ROC curve rises rapidly towards the upper-left-hand corner of the graph, this means the false-positive and false-negative rates are low. We may say that the diagnostic test is good. A bad diagnostic test is one where the only cutoff values that make the false-positive rate low have a high false-negative rate and vice versa. So if the ROC curve declines from the lower-left-hand corner to the upper-right-hand corner, the related diagnostic test might not be good.
  • Area under the curve
    If the ROC curve rises to the upper-left-hand corner, the larger the area under the curve, the better the diagnostic test. In practice, a diagnostic test is going to have an area somewhere between these two extremes. The closer the area is to 1.0, the better the test is, and the closer the area is to 0.5, the worse the test is.

Handling Missing Values

The missing values in the data range will be excluded in the analysis.

The missing values in the grouping range and the corresponding data values will be excluded in analysis

Performing ROC Curve

To perform a ROC Curve analysis:

  1. Select Statistics: ROC Curve. This opens the ROCCurve dialog box.
  2. Specify the Input Data and set Computation Control options.
  3. Upon clicking OK, an analysis report sheet is generated.

Topics covered in this section: