How do you interpreted level variance logit multilevel?

[Jose Saez-Lozano]

I have specified a multilevel logit model with three levels. I have specified the constant as a random effect both at level 2 and at level 3. I have also specified a variable X with random effects at level 2. The estimated coefficients are in log-odds.

The variance of the level 2 random effect is Var (U0jk + U1jk * Xijk). The variance of the level 3 random effect is Var (V00k)

1.- How do you interpret Level 2 Variance (U0jk) = 1.017?

2.- How do you interpret Level 2 Variance (Xijk) = 1,030

3.- How do you interpret Level 3 Variance (V00k) = 1.066?

Jose L

[Clyde Schechter]

Your question would be clearer if you posted the actual Stata command you used for the multilevel mlogit model and the actual output you got from Stata. (The direct output, not laundered through -estout- or -esttab- or any other pretty-print program.) Different people assign numbers to levels in opposite orders, so it is guesswork as to what you are talking about here.