Dear Owen, To save the simulated quantities of interest, use the genpr(), genev(), or genpv() options associated with simqi. For example, genpr() saves the simulated probabilities, conditional on the x's you chose with setx, as new variables in the current dataset. Each new "observation" represents one simulated probability. You asked why the simulated probabilities are "different in each case." The main reason is estimation uncertainty. The parameter estimates are uncertain, so anything calculated with them -- including probabilities, predicted values, and expected values -- are also uncertain. Clarify takes that uncertainty into account. It combines the values you chose at the setx stage with simulations of the parameters to produce simulated probabilities (given the x's you chose, not the values in your dataset). Hope this helps. Best of luck with your research. Michael Tomz, Assistant Professor Phone: 650-725-4031 Department of Political Science Fax: 650-745-2765 Encina Hall, Stanford University tomz(a)stanford.edu Stanford, CA 94305-6044 On Tue, 14 May 2002, Owen Abbe wrote: clarify mailing list served by Harvard-MIT Data Center List Address: clarify(a)latte.harvard.edu Subscribe/Unsubscribe: http://lists.hmdc.harvard.edu/listis.cgi?info=clarify