Survey and Big Data Interactions

World Statistics Day: On the Importance to Society of High Quality Public Statistics – Joint IAOS-IASS Webinar 

IASS Webinar:

11am CET on September 3rd, 2020

Measuring the socio-economic impact of coronavirus in Asia and the Pacific. The coronavirus has been a major wake-up call for national statistical organisations. Survey operations have been impacted, new data sources are being used, data gaps are being filled by whoever is able to fill them, and best practices are being tested. In Asia and the Pacific, nearly 75% of National Statistical Offices have had to postpone fieldwork for planned censuses and there is thirst for information on what to do and how to do it. ESCAP, the UN Economic and Social Commission for Asia and the Pacific, has collated information on socio-economic surveillance surveys being undertaken in Asia and the Pacific and convened several dialogues to share expertise and experiences on these activities. In this presentation, ESCAP will share their findings and the common questions being asked by countries in relation to these surveys.

Overview of the design and estimation methods in the ONS Covid Infection Survey. ONS has been running a survey to measure prevalence and incidence of Covid-19 infection in the household population since April 2020. It was initially limited to England but it has now been extended to the whole UK, and the sample is being increased substantially with the aim to achieve 150000 responding individuals in a fortnight by October 2020. I will describe the challenges we have encountered in collecting and processing the data, as well in estimation and analysis.

Statistical Design for Covid-19 monitoring and control. The Covid-19 pandemic has affected countries differently. In New Zealand all incoming international travellers are put in isolation for two weeks and any cases found are quarantined. There has been limited community transmission. Contact tracing of community cases remains exhaustive. There is a Statistical Advisory Group to the NZ Ministry of Health. The underlying strategy has been elimination rather than eradication, via a scale of alert levels that utilise lockdowns and bubbles.

Although the NZ situation is not always replicated elsewhere where Covid-19 prevalence is higher, there remain common underlying statistical themes and issues. • The need for government Ministries and Departments of Health to prepare by commissioning design of prevalence surveys as soon as possible, even if implementation is delayed. • Recognition of the potential to integrate sound sampling with contact tracing by using repeated adaptive cluster and network sample designs, which make it possible to track all contacts of known cases, have a chance of detecting unknown community cases, monitor special groups (such as those at the border) and get prevalence estimates with standard errors over time. • The need to understand better the effect of the survey design on specificity of lab tests. • To recognise how pooling laboratory tests using even simple experimental designs could improve test sensitivity.

Gemma Van Halderen is Director of the Statistics Division in the United Nations Economic and Social Commission for Asia and the Pacific (ESCAP). Prior to joining the United Nations, Gemma was a member of the Australian Bureau of Statistics (ABS) Executive Team focusing on data sharing, data integration and microdata access and ABS’ contribution to the Australian Government’s Data Integration Partnership for Australia. In 2017, Gemma was seconded to the Commonwealth Department of the Prime Minister and Cabinet to lead preparation of the Government’s response to the Productivity Commission Inquiry into Data Availability and Use. Gemma has extensive global experience. She was inaugural co-chair of a UN Expert Group on the Integration of Statistical and Geospatial Information, is an elected member of the International Statistics Institute, a member of the International Association of Official Statistics (IAOS) Executive Committee, the regional editor of the Statistical Journal of the IAOS and an advocate for young statisticians. Gemma holds a Bachelor of Science with Honours from the Australian National University.

Salah Merad has been at the Office for National Statistics for over 10 years. He had led sampling and estimation development projects for several ONS surveys, including the Labour Force Survey and the Opinions and Lifestyle Survey, and the recent Covid-19 Infection Survey.

Stephen Haslett is Emeritus Professor of Statistics at Massey University New Zealand, Professorial Fellow at the National Institute for Applied Statistics Research Australia (NIASRA) at the University of Wollongong, and Visiting Fellow at the Research School of Finance, Actuarial Studies and Statistics (RSFAS) and former Professor / Director of the Statistical Consulting Centre at the Australian National University. He has theoretical interests in linear and mixed linear models, sparse contingency tables, sample design and analysis, and small area estimation. He has undertaken extensive small area estimation and survey design projects for the UN for poverty estimation, food security, nutrition and health, often related to emergency situations. Linked principally with these projects, he has worked in or for central government statistical agencies in more than two dozen countries mainly in Asia, Africa and the Pacific. He is currently involved in research at the Centre for Public Health Research in New Zealand and is a member of the NZ Ministry of Health (MoH) Statistical Advisory Group which is providing statistical advice for Covid-19 to MoH. He began his professional career during the late 1970s, designing sample surveys at Statistics New Zealand.



IASS Webinar:



Other Links

International Statistics Institute [YouTube]