Strategies for eliciting and merging confidential social research data

This report presents, discusses and extends strategic models for representing the process of merging records from different sources when confidentiality of the records is required by law or custom. Examples and variations on the models cover simple situations, such as eliciting anonymous data from previously identified respondents, as well as more complex merge operations, such as merging files from different data archives and merging data under code linkage systems. The versatility and potential corruptibility of the models are also discussed.Supported by NSF Grants GI29843 and GS320-73X. Professors Eli Rubenstein, Donald T. Campbell, and Lance Hoffman were generous in providing advice on earlier drafts of the paper. However, views expressed here do not necessarily reflect the views of the sponsoring agency or of these individuals.