On Thu, 25 Mar 2004 11:29:54 -0500, Yucel, Recai M. <[email protected]> wrote:
> One solution is to wait for the stand-alone version of PAN. Second is to > get > the earlier versions of Splus. If these are all impractical, you can use > other packages under assumption that your sample is drawn under simple > random sampling scheme. There are several such packages, one of them is > NORM. You could introduce a "cluster" dummy in your models. You should be > cautioned, however, that the standard errors will be artificially small. > > Recai Thanks. I got pointed off-list to R, and the PAN port for it, which should do what I need. Thanks, all. Pat -- Patrick S. Malone, Ph.D., Research Scholar Duke University Center for Child and Family Policy Durham, North Carolina, USA e-mail: [email protected] http://www.duke.edu/~malone/
