Hello,
I am looking to examine the IRR using Kappa statistic in a situation where I believe that Fleiss' kappa is most appropriate, but I am unsure how to proceed in Stata and was looking for advice. Here is the scenario:
130 subjects receiving physical examinations
Each subject receives exams at identical time by 2 different raters completing the same exam report form
45 physicians total participated in these exam dyads over the course of the study period (ie., it was not the same 2 raters for every single individual who received an exam)
Looking to calculate interrater reliability of individual physical examination findings using these paired exams
I don't think that Cohen's kappa is best, since there are more than 2 raters participating in the overall study, although 2 raters provide data for each individual patient. Are there any suggestions on how to most appropriately analyze this in Stata? Appreciate any help and advice.
I am looking to examine the IRR using Kappa statistic in a situation where I believe that Fleiss' kappa is most appropriate, but I am unsure how to proceed in Stata and was looking for advice. Here is the scenario:
130 subjects receiving physical examinations
Each subject receives exams at identical time by 2 different raters completing the same exam report form
45 physicians total participated in these exam dyads over the course of the study period (ie., it was not the same 2 raters for every single individual who received an exam)
Looking to calculate interrater reliability of individual physical examination findings using these paired exams
I don't think that Cohen's kappa is best, since there are more than 2 raters participating in the overall study, although 2 raters provide data for each individual patient. Are there any suggestions on how to most appropriately analyze this in Stata? Appreciate any help and advice.