I have a dataset where two blinded outcome assessors independently scored a specific task.
The task is scored based on four different (dependent) parameters giving different points each: p1 (positive 0-10 points), p2 (positive 0-5 points), p3 (positive 0 or 2) and p4 (negative 0-2 points).
A total_score is calculated from these (= p1+p2+p3+(-p4)) and ranges from -2 to 17.
Data example:
Id. | p1_rater1 | p1_rater2 | p2_rater1 | p2_rater2 | p3_rater1 | p3_rater2 | p4_rater1 | p4_rater2 | total_score_rater1 | total_score_rater2 |
1 | 2 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 2 |
2 | 10 | 8 | 5 | 4 | 2 | 2 | 1 | 1 | 16 | 13 |
3 | 6 | 5 | 3 | 2 | 2 | 2 | 0 | 1 | 11 | 8 |
4 | 3 | 3 | 1 | 1 | 0 | 0 | 0 | 1 | 4 | 3 |
5 | 9 | 9 | 4 | 4 | 0 | 2 | 0 | 0 | 13 | 15 |
If possible and preferably, I want a test statistic that takes into account that the total_score is calculated from 4 different parameters.
Hope you can help.
Kind regards.
0 Response to Inter-rater correlation
Post a Comment