Skip to main content

Latest Tweets

Release of SIPP Synthetic Beta, version 5.1

Print Friendly, PDF & Email

We are pleased to announce the release of the SIPP Synthetic Beta, version 5.1 on the Synthetic Data Server at Cornell University. This version contains modeling improvements and new SIPP variables that expand the number of analyses that can be performed.

In particular the Census Bureau has added SIPP monthly time series variables for weeks with a job, weeks with pay, usual hours worked, survey-reported earnings, total personal income, any health insurance coverage, and employer-provided health insurance coverage.

The Census Bureau has also added two often-requested indicators: first, an indicator for state of residence at the beginning of the SIPP panel; second, an indicator for whether the individual linked to administrative records via an SSN or whether these records were imputed because no SSN was available.

Finally, the Census Bureau has edited the administrative earnings variables prior to the data completion and synthesis process in order to modify some values that we determined to be clear data error.

For more information on the contents of the SSB, see the Codebook for Version 5.1.
Users with accounts on the Synthetic Data Server will find the new data in /rdcprojects/co00517/SSB/data/v5.1/. The format of the files and the names of all the previously existing variables are the same as in version 5.0 so any previously written programs should still run.

While the format remains the same, it is important to be aware that version 5.1 contains new draws from the estimated distribution. Therefore, while any code written for version 5.0 should still work, results will differ slightly.

As of May 13, 2013, the directory /rdcprojects/co00517/SSB/data/current contains links to version 5.0 data which are stored in /rdcprojects/co00517/SSB/data/v5.0. As of July 23, these links were changed so that they point to /rdcprojects/co00517/SSB/data/v5.1/. Users wishing to continue using version 5.0 will need to change their directory names to explicitly reference the old data.

As with version 5.0, the Census Bureau continues to offer the opportunity for analysts to have their code validated on the underlying completed Gold Standard files. This process is unchanged. Analysts should write code using the SSB and test it on the Cornell SDS server. Once the code for a single implicate is finalized and runs without error, the analyst may request validation by emailing sehsd.synthetic.data.use.list@census.gov to request the appropriate disclosure review memo and provide the name and location of the program to be run by Census staff. For more information on this process, please visit http://www.census.gov/sipp/synth_data.html.