Newshub - NUS' News Portal

Tapping administrative data for policy formulation

13 August 2012

Prof Ho giving the opening address at the conference

A conference on "Evidence-based Public Policy Using Administrative Data", organised by NUS Department of Economics and the Civil Service College (CSC) and the National University of Singapore (NUS), brought international and local academics together with Singapore public policymakers to explore using administrative data to make better informed policy decisions backed by robust scientific evidence.

The first such meeting hosted jointly by a government agency and an academic institution, the two-day event was held early August at the Civil Service College.

Government agencies collect administrative information as part of functional or operational activities. Such data on individuals making up big sample sizes offers opportunities for research into evaluating and formulating public policies in international trade, education and family, labour and behavioural economics, as well as health and development.

At the conference, Guest-of-Honour Professor Ho Teck Hua, NUS Vice President (Research Strategy) and Tan Chin Tuan Centennial Professor, welcomed the some 280 attendees from the government and educational institutions.

The keynote presentations by Dr Victor Lavy (William Haber Chaired Professor of Economics at Hebrew University of Jerusalem) and Dr Zhang Junsen (Wei Lun Professor of Economics, The Chinese University of Hong Kong) talked about research based on administrative data showed how expanding the budget allocated per class of students could positively influence academic achievement and behaviour in Israel; and how population control policies affected household saving patterns in China.

NUS faculty staff who presented included Department of Economics Associate Professor Wong Wei Kang who asked whether an additional year of schooling helps children in improving their reading, mathematics and science; and School of Business research fellow Mr Sadat Reza who looked at driving the world's most expensive cars.

Dr John Abowd, Edmund Ezra Day Professor of Economics of Cornell University, rounded up the plenary session by examining the statistical issues that arose in the design of the Census Bureau's Longitudinal Employer-Household Dynamics infrastructure file system and its associated public-use products. A panel discussion on the considerations and options of setting up administrative data access saw various inputs and opinions from participants, providing nuggets for further thoughts and strategies for those present.