MIssing library: Appropriate model for binary items

[Shawn Boles <shawn@ori.org>]

Hi All:

The support materials for the missing library (S+ 6.01 r 4 on Solaris 5.6)
are silent as to whether one should adopt a Gaussian (continuous) or a
loglinear  (factor) approach to imputation of binary items.  I would
appreciate any advice from those who may have experience in using the
library with binary items.

Thanks in advance,

Shawn Boles, Ph.D.
Senior Research Associate
Oregon Research Institute
1715 Franklin Blvd.
Eugene, Oregon 97403-1983
USA
Phone (541) 484-2123 ext 2225
Fax: (541) 484-1108