I have a dataset of 50 annotators that highlight(codified as 1) or not (codified as 0) a text, according to 2 different rules (call them A and B). The dataset is divided into a column for each word of the text. Here a simplified example of the database:
I would like to compare the agreement between the raters in A and raters in B for each word and see if there is an overlapping in what the annotators highlighted, even if the instructions were different. I am quite new to using R and I am looking for the best way to proceed in these kinds of analysis. I have considered Krippendorff's alpha, but it supports only one variable at a time. Furthermore, I want to have info for each word of the text, and not a general value. Thank you in advance for your help.