A disclosure would occur if someone could infer data values (e.g. dollar value of sales) for a particular business that has provided information via a census or survey form under a pledge of confidentiality.
The Census Bureau is committed to confidentiality and constantly pursues new procedures, technologies and methodologies to safeguard individual data.
Disclosure avoidance is the process for protecting the confidentiality of data. The Census Bureau uses two methods of preventing disclosure of business data, cell suppression and noise infusion.
Some of the data in economic census tables are withheld to protect the confidentiality of information reported by individual businesses. Data withheld are replaced with ‘D’s in appropriate data cells.
Cell suppression protects the confidentiality of individual businesses by cell values in tables where the amount of the cell, if it were known, would allow one to estimate a single contributor’s value too closely. This occurs when there are very few contributors or when there are one or two large contributors that dominate the aggregate statistic.
The cells that must be protected are called primary suppressions.
To make sure the primary suppressions cannot be closely estimated by subtracting the other cells in the table from the higher-level totals, additional cells may also be suppressed. These additional suppressed cells are called complementary suppressions.
The process of suppression does not usually change the higher-level totals. Values for cells that are not suppressed remain unchanged. Before the Census Bureau releases data, computer programs and analysts check published tables for both primary and complementary disclosures.
Establishment counts are not considered to be disclosures, so this information is published in all tables.
Ranges are sometimes used in place of ‘D’s to suppress sensitive data, but still provide meaningful information.
Noise infusion is an alternative to cell suppression that allows for the publishing of more data. By marginally adjusting (perturbing) each respondent's data, data for individual businesses can be camouflaged. Most of the resulting aggregated statistics are distorted by a relatively small amount. Some cells may be suppressed for additional protection from disclosure or because the quality of the data does not meet publication standards. Though some of these suppressed cells may be derived by subtraction, the results are not official and may differ substantially from the true estimate.
Protective noise is applied to the following economic tables:
For more technical information on noise infusion, read "Using Noise for Disclosure Limitation of Establishment Tabular Data" [PDF 105KB] by Timothy Evans, Laura Zayatz and John Slanta in the Journal of Official Statistics (1998).
The Census Bureau is bound by Title 13 and Title 26 of the United States Code. Title 13 provides the authority to conduct censuses and surveys, and both Titles 13 and 26 provide strong protections for information collected from individuals and businesses.
Other federal laws, including the Confidential Information Protection and Statistical Efficiency Act [PDF 80KB] and the Privacy Act, also reinforce these protections.