In the case of the kappa-value there are some attempts to qualify how good or bad the agreements are. For example Landis & Koch in the article The Measurement of Observer Agreement for Categorical Data talks about "strength of agreement" based on kappa values:
Kappa Strength of agreement ===== ======================= 0.0-0.20 Slight 0.21-0.40 Fair 0.41-0.60 Moderate 0.61-0.80 Substantial 0.81-0.90 Almost perfect
My question is if there are some attempts to qualify the strength of agreement" based on "quadratic weighted kappa values" in the same way. Any references about attempts to define "generic strength of agreement" using quadr. weighted kappa?
My assumption is that it would not be meaningful to define "strenght of agreement" based on kappa and to use the same qualification using quadratic weighted kappa. They are different values.