This paper proposes a method of designing a statistically distinguishable rating scale that is not excessive in relation to the existing observation statistics. This allows for more stable validation ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback