I have a small doubt regarding the comparison of two ratings.
I have two candidates, I asked them to rate themselves out of 10 in 4 different subjects. Both have the same level of knowledge in both subjects.
Student name Rating (Maths Physics Chemistry Biology)
A 10 10 10 10 B 7 7 7 7
(According to A he is giving 10/10 for his best skill, and B is giving 7/10 for his best skill.)
If I am comparing A and B in an absolute scale A is better than B. Is there a way or approach to detect them having the same skill? The only reference I have is the information in the table which is relative and I know they have the same skillset.
Please help me out, Correct me if I am wrong.
Thanks and regards, Sarin
With self-ratings? I don't see any realistic way to use them to compare between different students. There's some credibility in comparing how a student rated themself in different subjects, but across different students? You have no way of telling what standards they're using. Also, all sorts of social effects apply.
That said, student A is almost certainly lying. Whether he's lying to you to get attention, or lying to himself that he doesn't have more to learn, I wouldn't trust those ratings at all. (Why did I switch to gendered pronouns? See social effects above.) With just these numbers to work with, and no other information about the candidates, I would rather work with student B.
And all this? It's more social science than anything. The mathematics of it doesn't really come in until we have some actual credible data.