Dear Acexx
To calculate cross-level correlations, request the variance-covariance matrices of
estimates of fixed effects and variance-covariance parameters based on HLM2 by
checking the print variance-covariance matrices option in the Output Settings
dialog box accessed via the Other Settings menu. The keyword PRINTVARIANCECOVARIANCE
facilitates the same purpose in batch mode.
Let
r = number of random effects at level 1.
f = number of fixed effects
p = number of outcomes in a latent variable run
pm = number of alphas in a latent variable run
For HLM2:
* tauvc.dat contains tau (tau(pi)) in r columns of r rows, the next r2 lines are the
tau(beta), and then the inverse of the information matrix (the standard errors
of tau[s] are the square roots of the diagonals).
* gamvc.dat contains the gammas and the gamma variance covariance matrix.
After the gammas, there are f more rows of f entries containing the variancecovariance
matrix.
* gamvcr.dat contains the gamma and the gamma variance covariance matrix
used to compute the robust standard errors. After the gammas, there are f rows
of f entries containing the variance covariance matrix.
Once you have gamvcr.dat, you get the correlations between any two variables X
(level-1) and Z (level-2) by taking the covariance associated with these two, and
dividing it by the square root of the product of the variances of X and Z. To get
the associated p-value, you will have to use a table of critical values for the r(n)
distribution where n is the number of level-2 observations.
If this note apprears unclear, please consult the HLM User guide
Good Luck
|