Home > Research . Search . Country . Browse . Small Grants

PSC In The News

RSS Feed icon

Lam discusses shifts in global population, past and future

Thompson says LGBT social movement will bring new strength in push for tighter gun control

Yang says devalued pound will decrease resources for the families of migrant workers in Britain

Highlights

Frey's new report explores how the changing US electorate could shape the next 5 presidential elections, 2016 to 2032

U-M's Data Science Initiative offers expanded consulting services via CSCAR

Elizabeth Bruch promoted to Associate Professor

Susan Murphy elected to the National Academy of Sciences

Next Brown Bag

PSC Brown Bags
will resume fall 2016

Michael R. Elliott photo

Addressing Disclosure Risk of Contextualized Microdata in Survey Design

a PSC Research Project

Investigators:   Michael R. Elliott, Daniel G. Brown, Trivellore Raghunathan, Kristine M. Witkowski

As the contracting institution for this project, staff at the University of Michigan will be performing the following administrative, data processing, analysis, and writing tasks for this project.
1) Organize and host three 2-day meeting of advisory board, one 1-day meeting of disclosure review board, and one 2-day meeting of project team, as well as compile and distribute materials for these events.
2) Conduct reviews of the disclosure and survey methodology literature.
3) Construct small-area estimates of human populations.
4) Construct approximate contextual measures for disclosure assessments (using data provided by Iowa?s GSS Pilot survey and the data accessed through Michigan Census Research Data Center, MCRDC).
5) Develop methods and measures for three different assessments of disclosure risk, analytical utility, and disclosure survey costs.
6) Conduct assessments of disclosure risk and analytical utility for the 2010 and 2012 GSS Pilot data as well as for survey data provided through the MCRDC at the University of Michigan (if time allows and there is interest, conduct third assessment of disclosure survey costs).
7) Develop and validate models used in the prediction of survey respondent characteristics (using data provided by Iowa?s GSS Pilot survey).
8) Formulate sampling designs used in simulations (Michigan as lead, with Iowa assisting).
9) Create synthetic contextual data used in simulations.
10) Formulate experiments and run simulations to gather estimates of disclosure risk, analytical utility, and disclosure survey costs for a large number of datasets that vary in their sampling and database designs.
11) Produce analyses of simulation data as requested by principal investigators.
12) Produce 8 to 9 papers (Michigan and Iowa as joint collaborators).
13) Present research at four national conferences over the last two years of the project.

Funding Period: 01/16/2012 to 11/30/2017

Search . Browse