Statistics for everyone: ch.19 Kappa statistics for agreement

easier R than SPSS with Rcmdr : Contents

ch.19 Kappa statistics for agreement

Kappa statistics analyzes the agreement of the two diagnostics.

Both of these diagnoses refer to those that come out positive/negative. If both diagnostics came out positive, if both came out negative, if only one was positive, and another was negative, a total of 4 values can be calculated by collecting the cases where only one came out negative. Enter its value in each cell.

The result shows the Kappa value and its 95% confidence interval. This is not judged by the value of p.

Interpreting these results is usually a common use of Landis and Koch’s interpretation (The Measurement of Observer Agreement for Categorical Data, 1977).

These are 0 to 0.2: Slight , 0.2 to 0.4: Fair, 0.4 to 0.6: Moderate, 0.6 to 0.8: Substantial, and 0.8 to 1: Almost perfect.

To calculate s ensitivity, specificity (specificity), or Kappa values, you need tabulated data. This menu makes it very easy.

First, select two norminal variables.

Next, select a few options:

This is done in a tabular format, which is called a ‘frequency table’ or a ‘contigency table’, just called a ‘table’.

Next Part is for Analysis of paired data

easier R than SPSS with Rcmdr : Contents

=================================================