We have developed a sample Predominant Purpose Statement (PPS) that can and should be used by projects that intend to access data at the RDCs, when such data is intended to be synthesized by the NSF-ITR project (that’s almost all LEHD related data plus the LBD).
- Here’s what it looks like:
- This research is partially supported by the National Science Foundation Information Technologies Research Grant SES-0427889, which provides financial resources to the Research Data Centers. The NSF-ITR grant is a partnership between the Census Bureau and the RDC community that seeks to increase the availability of public use micro-data from Title 13, Chapter 5 data sets by applying the technique of synthetic data to the production of the public use files. Synthetic data based on the confidential micro-data used in this project directly satisfy criterion 1 because they improve the usefulness of the Census product that has been synthesized by providing an additional public-use data resource based on that product. In order for synthetic data to be useful, they must protect the confidentiality of the underlying micro-data and they must be analytically useful. Specialized teams of Census and external experts are developing the synthesizers and analytic validity tests as a part of a variety of internal Census projects. This project will support those efforts by providing archives of the estimation results and programming sequences used to create the analysis files. This archive will be used to test the analytic validity of synthetic data created from the input files used for the project. In addition, the archived models may be used by the synthesizer as part of the probability structure used to create the distributions from which the synthetic data are sampled. All synthetic micro-data created in conjunction with this project will be separately disclosure reviewed by Census and by any other agency that is a custodian of the underlying micro-data before they are publicly released.