Gwet’s AC2 is another measurement of inter-rater reliability. The advantage of this approach is that it supports categorical (Gwet’s AC1), ordinal, interval and ratio type data and also handles missing data.
Topics
References
Gwet, K. L. (2001) Handbook of Interrater reliability. STATAXIS Publishing
Gwet, K. L. (2015) On Krippendorff’s alpha coefficient
https://agreestat.com/papers/onkrippendorffalpha_rev10052015.pdf
Thank you for the web page. I’m trying to calculate Gwet’s AC2 with two judges, but I can’t, it only works with three or more judges. Is it possible to run it with two judges? Thank you.
I will look into this shortly. What sort of ratings do the judges use? (categorical, Likert, numeric, etc.)
Charles
I have tried with categorical and numeric and it doesn’t work in either case. Thank you for your response.
The functions GWET_AC2 and GWET_SES should work with only two judges. GWET and GWET_SER require more than two judges.
Thanks for bringing this issue to my attention. I will make this clearer on the website.
Charles
I am looking into this issue further and will issue a new Real Statistics release in a day or two to make a correction to the GWET function.
Charles
Great. Thank you so much.
LB,
I have made the change necessary, but need to test it further. Sorry, but I will need another day or two.
Charles
No problem. Thank you so much.
Sorry to bother you again. Is the Gwet function with two judges already working? I don’t mean to rush you at all, I just want to know if it already works. Thank you very much for your effort.
Yes, it is already working. See Blog
Charles
I have downloaded and it works perfectly. Thank so much again for your fantastic work!
Dr. Good morning, is there any difference between the statistics AC1, AC2 and B?
Thank you
Dr. buenos días existe alguna diferencia entre los estadísticos AC1, AC2 y B?
Gracias
Hello Gerardo,
AC1 is simply AC2 restricted to categorical ratings.
I don’t know what B is.
Charles
Hello-
Thank you for this wonderful site!
I was wondering how to calculate Gwet’s AC1. I only see AC2 as an option for reliability. Do we just run the AC2 agreement table analysis after converting our categorical data into an agreement table?
Thank you!
Glad you like the website.
Yes, AC1 is the same as AC2 when restricted to categorical data. Just use the AC2 data analysis tool.
Charles