alderton@NPRDC.NAVY.MIL (David Alderton) (02/08/90)
Folks,
Anyone have a suggestion as to how I can take the square of a _type_
= 'std' from a type=cov data set? For those that helped in the past,
this has to do with computing coefficient alpha. I, with a lot of help,
wrote some code to compute alpha based on the mean interitem correlation
matrix. Well, it turns out that this is mostly adequate but tends to
somewhat over estimate the true alpha. However, if alpha is computed
based on the covariance matrix it is accurate (in SPSSx parlance I believe
this is the difference between alpha and the standardized alpha which is
generally higher).
So, the problem I'm having is this. I can get the sum of the covariance
matrix but I need to get the sum of the variances as well to to calculate
alpha = (k/k-1) * (1 - (sum_of_variances / sum_of_covariance_matrix)).
The type=cov data set contains the standard deviations (_type_='std')
but have not found a way to turn them into variances or to address just
the diagonal of the covariance matrix.
Any ideas?
Thanks,
David.
David L. Alderton, Ph.D.
Navy Personnel Research and Development Center
Aptitude Research Division
Code 131
San Diego, CA 92152-6800
(619 553-7647 or AUTOVON 553-7647)
arpanet: alderton@nprdc.navy.mil
=============================================================================
| The opinions expressed or implied are mine, are not official, and do not |
| necessarily reflect the views of the Navy Department or the US Government |
=============================================================================