Hi,
Is it possible to obtain a p value to so say how different 2 kappa values are from one another?
I am helping someone who has 2 seperate groups of patients (low risk and medium risk, n=52 in each group) , and within each group a number of measurements have been taken, including a score of 1 to 6 to quantify the amount of coronary stenosis. In both groups the score has been assessed by 2 seperate operators to estimate inter-observer agreement and 1 of the operaters has also repeated the scoring to assess intra-observer agreement (using Kappa).
In the inter-observer study the kappa values and their 95% CIs have been calculated and dont overlap so there appears to be a significant difference between the likelihood of agreement in low risk compared to medium risk patients. The researcher I am helping would like to put a p value to this difference in the levels of agreement, but I am not sure how to do this.
I would be very appreciative of any help people could give me.
Many thanks,
Michael
_________________________________________________________________
Who's friends with who and co-starred in what?
http://www.searchgamesbox.com/celebrityseparation.shtml
|