FYI, the new issue of The American Statistician includes a review that points out biases in SPSS 12.0 Missing Values Analysis (MVA) software: http://www.sociology.ohio-state.edu/people/ptv/publications/MVA/published.pdf Researchers who are using or considering MVA will be interested in this review. Other readers may be interested in the examples, which demonstrate simply, in closed form, potential biases in naive missing-value methods. These examples may prove useful in the classroom.
Best wishes, Paul von Hippel Paul von Hippel Department of Sociology / Initiative in Population Research Ohio State University 300 Bricker Hall 190 N. Oval Mall Columbus OH 43210 614 688-3768
