Lead Supervisor: Natalie Shlomo
Contact Email: Natalie.email@example.com
Partners: University of Manchester
External Partners: ONS
Start Date: September 2019
The Administrative Data Research Programme led by Hannah Finselbach at the ONS have identified a series of tasks in a preliminary work plan to advance the use of administrative data in official statistics. Table 1 contains a summary of tasks based on the preliminary work plan. The final work plan will form the basis for a strategy document that will facilitate research and engagement with the academic community.
Working with the ONS, the project team (the student and supervisors) will select a set of tasks that are appropriate topics for PhD research in advancing the use of administrative data in official statistics.
This is an exciting opportunity for a forward thinking doctoral student with a background in social science, statistics or computer science, to build a research profile in an important emerging area.
Table 1: Summary of tasks on advancing the use of administrative data in official statistics which will inform the PhD research questions
|Quality||Measuring and communicating uncertainty of outputs including quality measures/intervals for estimates using administrative data and developing the ‘Total Administrative Data Error’ framework
|Linkage||Linking multiple sources with possible different hierarchies, developing methodology and tools and measures of quality for linked data
|Privacy and Confidentiality||Generating synthetic data and assessing disclosure risk and data utility, new forms of data dissemination, investigating the potential of differential privacy
|Estimation||Comparison of administrative data sources with survey data and identifying and addressing under or over-coverage issues using case studies and applications
|Statistical Data Editing
|Editing and imputation, data cleaning and harmonization of administrative data sources
|Transactional Data||Transactional (streamed) data and dealing with time lags and combining aggregate data using time series approaches
Deadline 7th April 2019
Reference number MN33