In record linkage, the correct statistical model underlying a particular application may present estimation difficulties. Often, a convenient model is substituted in place of the correct one. Naturally, the substitution induces an error and one can only hope that the error is negligible. This paper compares two models as they are applied to the data collected during the 1988 Dress Rehearsal of the Decenial Census. Record linkage Decision rule Conditional independence model.
This article is from 'Computing Science and Statistics: Proceedings of the Symposium on Interface Critical Applications of Scientific Computing (23rd): Biology, Engineering, Medicine, Speech Held in Seattle, Washington on 21-24 April 1991,' AD-A252 938, p415-418.