presentation - VirtualRDC @ Cornell - Cornell University
presentation - VirtualRDC @ Cornell - Cornell University
presentation - VirtualRDC @ Cornell - Cornell University
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
Item suppression<br />
■ Some disclosure risk remains for counts based on very few<br />
entities in a cell (fewer than three individuals or employers)<br />
■ Variables affected are: B, E, M, F , A, S, H, R, FA, FH,<br />
FS, JC, JD, JF , FJC, FJD, FJF .<br />
➜ item suppression based on the number of either workers or<br />
the number of employers that contribute data for that item in<br />
a cell k in time period t, where a cell represents a particular<br />
combination of geography × industry × age × sex.<br />
■ Because of noise infusion, no complementary suppressions<br />
are needed<br />
■ Some denominators may be zeroes - the ratio or rate cannot<br />
be computed.<br />
August 10, 2005 - p. 14/31