There are two binary raters:
Rater 1 is certified to assess the ability of people to drive a taxi.
Rater 2 is certified to assess the ability of people to drive a lorry.
Rater 1 and 2 have both assessed the same cohort in City A for driving taxis and lorries, respectively.
In City B, we need to assess the ability of people to drive lorries, but only Rater 1 can come to City B.
In order to know to which extent Rater 1 can assess for lorries, how useful/meaningful would be to use Cohen's kappa for the agreement between Rater 1 and Rater 2 on assessing "the ability to drive either taxi or lorry" (both targets merged) obtained from City A results?
Could this coefficient be used to evaluate the reliability between raters who have different but related aims/targets?