The Statistical Society of Australia and AMSRS are proud to present the following workshop:
Improving inferences from poor quality samples
Melbourne, 15 August 2017
presented by Paul J. Lavrakas, Dina Neiger and Darren Pennay
About the workshop:
For many years now probability surveys based on random sampling of the in-scope population where every member of that population has a known, non-zero chance of selection have represented the gold standard for survey research in fields such as public opinion, behavioural and marketing research.
Nowadays, however, attainment of this ‘gold’ standard is more aspirational than realistic. This is due to factors that have eroded our ability to achieve this gold standard as well as to the rise of alluring timely and relatively low cost alternatives. The two main factors that have contributing to making probability-based sampling of the population more difficult and more expensive are the greater difficulty in accessing population frames that reliably cover the population of interest (i.e. non-coverage) and the precipitous decline in response rates in recent years (i.e. non-response). The most of alluring and popular of the alternatives to using robust probability based surveys to measure population-based phenomenon are non-probability ‘opt in’ online panels and big data analytics.
In the past survey research has proven to be resilient and adaptable in the face of such challenges and demonstrated that well designed and well executed probability based surveys of the population, even those with relatively low response rates, are capable of producing accurate population estimates. The questions this workshop helps address are:
- What is the current situation with respect to the accuracy of the estimates obtained from probability-based surveys?
- How achievable / realistic is gold standard nowadays?
- Given the difficulties in undertaking high quality probability surveys what are survey researchers and statisticians doing to improve the accuracy of the survey estimates we generate from poorer quality and non-probability samples.
This workshop will adopt a solutions-based approach and outline techniques, illustrated by case studies, which show how inferences can be improved from surveys administered to biased, low response rate and non-probability samples.
Topics covered will include:
- Declining response rates over time,
- Advances in survey execution and weighting techniques to adjust for non-coverage error and reduce non-response bias (i.e. non-response surveys, use of auxiliary data, adaptive and response design (including response propensity modelling), etc.,
- Advances in techniques to improve estimates from non-probability samples illustrated by case studies in the areas of:
- Sample blending
- Model-based design weights.
About the presenters:
Paul J. Lavrakas, PhD, is a research psychologist and research methodologist, and during the past four decades has conducted research, written many articles and books about research, and taught about social research methods. He joined the NORC at the University of Chicago in 2014 as a Senior Fellow, where he is heavily involved in helping to create its new probability-based online panel, AmeriSpeak.His work at NORC also involves devising experiments to raise response and consent rates, and to lower total survey costs.
Paul has a strong connection to Australia (with his son’s family living in Perth) and has been a regular visitor to Australia since 2002. Paul’s Australian roles include working as a Senior Methodological Adviser to the Social Research Centre and being a member of the Scientific Advisory Board for the ANU Centre for Social Research and Methods.
In previous roles Paul has consulted for various public and private sector organizations, including Google, The Associate Press, the Inter-American Development Bank, the Media Ratings Council, and Stanford University.
He is a Visiting Scholar at Northern Arizona University, where he teaches a methods course each year, and a Senior Research Fellow to the Office from Survey Research at Michigan State University. From 1978-2000 he was a Professor at Northwestern University and then Ohio State University, and was the founding faculty director for the survey centres at each institution. From 2000-2007 he served as chief methodologist for Nielsen Media Research.
Dr Lavrakas is a past President of the American Association of Public Opinion Research.
Dina Neiger, PhD, is a professional statistician with over 20 years of experience and a track record of achievement in leadership and technical roles at the Social Research Centre, Monash University, Australian Bureau of Statistics, and Biostatistics and Clinical Trials Centre at Peter MacCallum Cancer Centre. Dina’s statistical interests include the use of calibration and blending methods to improve accuracy of the non-probability samples, establishment and maintenance of the first Australian Online Probability panel and complex business survey design and weighting. Throughout her career she has worked through every stage of statistical data collection including design, system development, contact with respondents, data editing, estimation and output, in a wide variety of domains including demographic, labour, business and price index statistics. Dina’s educational background includes 1st Class Honours degree in Statistics and PhD in Business Systems from Monash University with an emphasis in applied Operations Research and Process Engineering. Dina is Accredited Statistician (AStat) member of the Statistical Society of Australia (SSA) and a full member of the Australian Society for Operations Research (ASOR).
Darren Pennay is the founder and Chief Executive Officer of the Social Research Centre. Darren is also an Adjunct Senior Fellow at the ANU Centre for Social Research and Methods and an Adjunct Professor with the University of Queensland’s Institute for Social Science Research (ISSR). Darren has worked in social research and survey design since 1984. He has previously worked with the Australian Institute of Multicultural Affairs, Australian Bureau of Statistics, Reark Research and the Wallis Consulting Group.
Darren formed the Social Research Centre late 2000 to develop a ‘centre for excellence’ in government and social research and to offer government, academic and not for profit researchers the choice of working with an agency specialising in this area of research. The Social Research Centre has grown from a single person start up entity to Australia’s leading supplier of research services to government and academic institutions.
Darren’s specific interests include large-scale survey design and management, longitudinal surveys, community attitudes research, health research, policy development research and survey methodology. He is a full member of the Australian Market and Social Research Society and a Society Fellow, has QPMR (Qualified Practicing Market Researcher) accreditation. Darren is also a member of the American Association of Public Opinion Research, the World Association of Public Opinion Research and the European Survey Research Association.
From 2010 to 2015, Darren played a leading role in the introduction of dual-frame telephone surveys to Australia. More recently he has been the driving force behind the establishment of Australia first probability-based online panel – the Life in Australia panel. Darren is also a leading proponent of the Total Survey Error approach to survey research. In 2014 he was awarded the Research Industry Council of Australia’s Research Effectiveness Award for Innovation and Methodology.
The registration fees are as follows:
SSA/AMSRS* Members: $ 220
Non-members: $ 330
The registration fee includes the following: workshop attendance, workshop materials, morning and afternoon tea as well as lunch.
*AMSRS members – Please register using the following link: https://www.amsrs.com.au/events/event/course17-sample
Registration closing date
7 August 2017 (registrations after this date will incur a late fee of $100)
Occasionally workshops have to be cancelled due to a lack of subscription. Early registration ensures that this will not happen. Please contact the SSA Office before making any travel arrangements to confirm that the workshop will go ahead, because the Society will not be held responsible for any travel or accommodation expenses incurred due to a workshop cancellation.
Cancellations received prior to Tuesday, 8 August 2017 will be refunded, minus a $20 administration fee.
From 8 August 2017 no part of the registration fee will be refunded. However, registrations are transferable within the same organisation. Please advise any changes to [email protected].
|Time:||10:00 am - 4:30 pm|
|Location:||Deloitte, Level 10,
550 Bourke Street,