Steve H. Cohen and Bogong Li (2006) "A Comparison on Data Utility between Publishing Fixed Intervals versus Traditional Cell Suppression on Tabular Employment Data"
The Bureau of Labor Statistics' Quarterly Census of Employment and Wages series publishes employment and wage estimates classified by county and six-digit North American Industry Classification System level. Unfortunately, due to data disclosure issues, nationally sixty percent of the county-level detail is suppressed using traditional cell suppression techniques. Alternative disclosure-protecting techniques have been proposed, specifically replacing these tabulations with fixedinterval for the suppressed cells. This paper will focus on comparing the data utility of these disclosure avoidance techniques. The positives and negatives of each will be discussed in addition to utility tradeoffs between the two methodologies. Data simulations will be presented.
Last Modified Date: January 9, 2007