Dealing with "Dirty" Data Is Topic of Westat White Paper for Chance Magazine


Westat’s Stephen “Shep” Roey, Tom Krenzke, and Robert Perkins, Jr., have co-authored a white paper in Chance that discusses how best to deal with “dirty” data, or inaccurate data, in a way that uses the most empirical methodological approach.

Masking data in order to protect the confidentiality of respondents is a key element in survey work. In Considering the Impact on Disclosure Risk as Illustrated in the Education Data Context, the authors warn that being overly cautious in masking dirty data can reduce data utility. They go on to offer solutions to consider for this problem.

Westat has far-reaching expertise in working with statistical disclosure control issues. We have worked with disclosure review boards for a number of Federal agencies, developed software, written disclosure analysis plans and papers, and conducted presentations and workshops on this topic.