Definition: p% rule

Data protection

This  rule states that a cell, contained in a table of magnitudes, is at risk if the value for any contributor can be calculated to within a given percentage. Cells that are identified as sensitive must be suppressed or otherwise avoided (through table design).
The p% rule calculates the distance (as a percentage) between the estimated value and the true value for the largest contributor in a table, as follows (see source for mathematical formula):

X is the value for the largest contributor, ^X = Total - Y is the estimate of X, and Y is the value for the second largest contributor. If the value of p is less than the p% threshold then the cell is deemed sensitive. If a cell is sensitive then aggregation or suppression (with secondary suppression) must be applied to the table. If a contributor’s value is negative, take the absolute value before calculating p.

The value of the threshold is confidential and is not to be made public.
Statistics New Zealand, “Microdata access output guide”, Wellington, New Zealand, October 2013

Search box