Disclosure Risk Assessment for National Education Data

• Activity Number: 666
• Type: Contributed
• Date/Time: Thursday, August 4, 2011 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistical Education
• Author(s): Yang Yang*+ and Xiaolei Wang and Nathan Menton and Jinae Boyer and Kyosin Kang
• Companies: ESSI and American Institutes for Research and ESSI and ESSI and ESSI
• Keywords: Disclosure risk ; Education data ; Confidentiality ; teacher characteristic ; school characteristic

Abstract

Federal statistical agencies are confronted with new challenges of maintaining confidentiality of education data, as state education agencies have increased the accessibility of school- and individual-level data through the Internet, and education marketing and sales companies have made various types of education data available for the general public to purchase. This paper examines the potential of disclosure risk of a national school and personnel dataset, by comparing this national dataset to a commercially available data file, which provides marketing information and services for the education markets. Data available for purchase includes teacher and principal characteristics, school programs and school characteristics. The analyses found that a number of variables were shared and that these shared data were reasonably consistent between the two databases. It was also found that a number of variables could be used to disclose the identity of the school or the personnel when the two data files are compared side-by-side. To maintain confidentiality of respondents, variables of potential disclosure risk need to be removed, combined with other variables, or recoded as categorical.