RC33 Eighth International Conference on Social Science Methodology

Monday July 9, 2012

Location:
University of Sydney Law School (Camperdown)
New Law School Building (F10)
Eastern Avenue, Camperdown Campus
The University of Sydney NSW 2006

For map click here

Room: L.L 100

Workshop presenters: Roy Lay-Yee and Barry Milne

Microsimulation is a methodological approach that is becoming increasingly relevant particularly but not only in the policy arena. It can draw together information from micro-level data, giving scope to human agency while taking into account social context. How does microsimulation fit with the social sciences, and what are its advantages? The workshop will attempt to answer these questions and introduce the general features of the microsimulation approach. As an example of a social science application, we will show the construction of our model  of determinants and outcomes in the early life course and how the associated tool can be used for testing policy scenarios. All welcome. No pre-requisites or technical expertise are required.

Room: L.L 107

Workshop presenter: Rob Ackland

This workshop will provide an introduction to social media network analysis using NodeXL (a free Excel 2007/2010 template for social network analysis, which has several plugins for social media data - http://nodexl.codeplex.com) and VOSON (http://voson.anu.edu.au, http://www.uberlink.com), which is a tool for hyperlink network construction and analysis, and is available both as a web application and NodeXL plugin (VOSON+NodeXL).
The workshop consists of three parts. Part 1 (1.5 hours) will introduce core SNA concepts as they relate to social media data and also basic NodeXL functionality. Part 2 (45 min.) will be a NodeXL demonstration using data from Twitter. Part 3 (45 min.) will be a demonstration of VOSON+NodeXL.
It is assumed that workshop participants will already be familiar with SNA and social media. While participants are welcome to bring own laptops and follow the demonstration, the workshop presenter will only be able to provide limited one-to-one assistance during the workshop. Furthermore, it will be assumed that participants who plan to use their own laptops have already successfully installed NodeXL and any necessary user accounts for access to data (e.g. Twitter, VOSON) have already been obtained.

Robert Ackland is an Associate Professor in the Australian Demographic & Social Research Institute at the Australian National University, where he conducts empirical social science research into online social and organizational networks. He leads the Virtual Observatory for the Study of Online Networks project (http://voson.anu.edu.au), coordinates the ANU's Master of Social Research programme and teaches on the social science of the Internet and online research methods. Robert has degrees in economics from the University of Melbourne, Yale University (where he was a Fulbright Scholar) and the ANU, where he gained his PhD in 2001.  He has been chief investigator on four Australian Research Council grants and in 2007, he was a UK National Centre for e-Social Science Visiting Fellow and James Martin Visiting Fellow at the Oxford Internet Institute.

Room: L.L 105

Workshop presenter: Adam Zammit

The queX Suite represents a new generation of survey software -- it is designed to simplify multi-mode surveys; it is web-based; it is open-source; and it can be installed locally or run “on-demand” on the cloud using a thin client.

It includes tools for survey data collection and coding in 3 survey modes (web, paper, CATI). The workshop will demonstrate the tools individually and how they work as a complete data collection suite: we will create an instrument in Limesurvey; deploy it as a scannable paper form (and using queXF to process completed questionnaires); administer the same instrument in queXS for CATI; and use queXC for data cleaning and coding/classification system.

The workshop will cover practical issues choosing whether to install locally or from the cloud; dealing with your IT department; and Dominique Joye from the University of Lausanne, will provide a user’s perspective on the software.

Room: L.L 102

Workshop presenters: Kimberly Fisher and Roger Patulny

Room: L.L 028

Workshop presenter: Joerg Blasius

It is well-known that survey data is plagued with non-substantive variation arising from myriad sources such as response styles, socially desirable responding, failure to understand questions, and even faked and/or duplicated interviews. In general one can say that all data contain both substantive and non-substantive variation. Modifying Box’s  (1987) famous quote that “[essentially] all models are wrong, but some are useful” I suggest that “all data are dirty, but nevertheless some are informative”. But what is “dirty” or “poor” data? The guiding rule is that the lower the amount of substantive variation, the poorer is the quality of the data. Applying principal component analysis, categorical principal component analysis and multiple correspondence analysis, I like to show various strategies for assessing the quality of the data; i.e., for detecting non-substantive sources of variation.

The workshop focuses on screening procedures that should be done prior to assessing substantive relationships. Screening survey data means searching for variation in observed responses that do not correspond with actual differences between respondents. It also means the reverse: isolating identical response patterns that are not due to respondents holding identical viewpoints. This can be a sign of faked and duplicated interviews.

In the workshop I will demonstrate a variety of data screening processes that reveal distinctly different sources of poor data quality. Using well-known data sets such as the ISSP and the World Value Survey, I will provide examples for how to detect non-substantive variation that is produced by:
Response styles such as acquiescence, extreme response styles, and mid-point responding;

• Misunderstanding of questions due to poor item construction;
• Heterogeneous understanding of questions arising from cultural differences;
• Different field work standards in cross-national surveys;
• Inadequate institutional standards;
• Missing data (item non-response);
• Respondent fatigue;
• Faked and partly faked interviews.

Room: L.L 022

Workshop presenters: John Maindonald and Matthias Ganninger

This will give an outline sketch of abilities that are available from R and from R packages.   Areas that may be noted include Learning resources for R, R-related books, Data input, Graphics (including dynamic graphics), Data manipulation,Interfaces to databases and to the web,   Regression, Classification,Multivariate methods, Spatial methods,   Survey analysis, Textual analysis, and Network analysis.  The aim will give a sense of the wide range of abilities that are available in R, wherever possible using graphs to illustrate what can be done.    Questions will be encouraged.  Those who bring laptops, with R and relevant packages installed, will be able to reproduce at least some of the graphs on their own systems.  Details of packages that it will be useful to install can be found at

http://www.maths.anu.edu.au/~johnm/courseweb/acspri-R.html
This web page will be updated with further details closer to the time of the course.

Dr John Maindonald is a Visiting Fellow in the Centre for Mathematics and Its Applications at the Australian National University. His research interests include Statistical Computation, Statistical Perspectives on Data Mining, Use of the R system for Practical Data Analysis, Research Planning, and Population genetics. He is the author of 'Statistical Computation' (Wiley 1984), and the senior author of ‘Data Analysis and Graphics Using R: An Example-Based Approach’ (Cambridge University Press, 3rd end, 2010).  He has co-authored papers widely in many different areas of biological and social science.

Dr. Matthias Ganninger is researcher at the GESIS – Leibniz Institute for the Social Sciences in Mannheim, Germany. For his doctorate from the University of trier, he has worked on design effects and published many articles in internatio- nal scientific journals. He co-authored a textbook on data analysis in the social sciences. He has also taught specialized courses on survey methodology in Germany and elsewhere in Europe.

University of Sydney, Quadrangle Cloisters and Lawns

Tuesday July 10, 2012

Location:
University of Sydney Law School (Camperdown)
New Law School Building (F10)
Eastern Avenue, Camperdown Campus
The University of Sydney NSW 2006

For map click here

Wednesday July 11, 2012

Location:
University of Sydney Law School (Camperdown)
New Law School Building (F10)
Eastern Avenue, Camperdown Campus
The University of Sydney NSW 2006

Location: Room L.L 106

While traditional survey research faces problems of declining response rates, diminishing land line coverage, and increases in controlled access problems, at the same time people are increasingly using social media to actively share volumes of personal information publicly, online & via their smartphones. Several emerging data capture and analysis methods take advantage of the public’s willingness to share personal information and participate in online networks with complete strangers, and as such, offer the potential for enhancing survey data collection. This plenary session will provide an overview of conducting research with these technologies, including infoveillance and sentiment analysis of online conversations, crowdsourcing response to surveys, and conducting surveys within Facebook, Twitter and Second Life. 

Elizabeth Dean, a survey methodologist, has 14 years of experience designing and pretesting survey instrumentation. She has published research on a variety of innovative applications of survey methodology, including conducting cross-cultural questionnaire appraisals, conducting surveys in virtual worlds, improving usability test methods, and using cognitive interviews to test consent forms.

Location: Room L.L 026

Most social science research is carried out in natural settings. Yet such research can rarely generate inferences of a plausibly causal status sufficient to inform social interventions. However, advances have been made in helping researchers develop and draw more credible inferences from such data. These advances have come particularly from logicians and philosophers, who have generalised to observational work a variant of the model of causal inference based on the experiment (potential outcomes, counterfactuals) and from applied statisticians, particularly those working in econometrics and in educational and applied social research who are concerned with drawing conclusions about policies and interventions. The plenary will review this work - which is preparatory to compiling a reader - and will also ask the question whether this model of causal inference can also help the social sciences build theory.

Peter Davis is Professor of the Sociology of Health and Well-Being at the University of Auckland, with cross-appointments in Statistics and in Population Health. He trained in sociology and in statistics at the London School of Economics, and has a PhD in community health from Auckland. He is also Director of the COMPASS research centre which focuses on health and social policy applications using advanced methodological techniques, mostly with existing data. He is currently Senior Editor, Health Policy, at Social Science and Medicine.

Location: Room L.L 024

The past, present, and future development of the total survey error (TSE) paradigm is discussed. Particular attention is given to the application of the TSE perspective to comparative research, especially cross-national research. Among the points to be covered are the interactive nature of much of survey error across its many different components and how the TSE approach can help to achieve functional equivalence in cross-national research.

Tom W. Smith is an internationally recognized expert in survey research specializing in the study of societal change and survey methodology. Since 1980 he has been a principal investigator of the National Data Program for the Social Sciences and director of its General Social Survey (GSS) at the National Opinion Research Center, University of Chicago.

Smith was co-founder and Secretary General of the International Social Survey Program (1997-2003). In 2011-2012 he is President of the World Association for Public Opinion Research.

Location: Room L.L 104

Carbon emission mitigation is one of the most pressing environmental challenges of our time. Policymakers worldwide are responding with carbon pricing mechanisms, but are increasingly looking to mechanisms beyond price/regulation. One such approach is the notion of “nudging” (Thaler and Sunstein 2008), which requires the design of a choice architecture to help consumers make “better” choices (form a policy perspective). For nudging to be successful, however, policymakers need knowledge of consumer behaviour (to be effective policies must be consistent with the nudges of the marketplace). This exploratory study investigates five emission mitigation strategies that consumers can adopt: energy efficiency, moderation, renewables, offsets, and recycling. Rather than question consumers directly on their preferences for the strategies, the researchers observe consumers’ choices of carbon-intensive household products, specifically, air-conditioners (A/Cs) and refrigerators (Fridges). The attributes of A/Cs and Fridges embed the emission mitigation strategies. For example, consumers can purchase a smaller refrigerator (moderation) with a more efficient compressor (efficiency). The analysis of these choices is based on conditional and mixed logit models, and structural logit models. These models provide insight into (1) consumers’ aggregate preferences, (2) unobserved sources of heterogeneity/variation in consumers’ preferences, and (3) the structure of this heterogeneity. A key finding is the remarkably similarity in the preference heterogeneity in consumers’ choices of A/Cs and Fridges. Guidelines or policymakers are offered based on this key result, especially for the design of initiatives targeted at changing the consumer behaviour of emission mitigation.
 
Len Coote is associate professor of marketing and discipline leader (marketing) in the UQ Business School, University of Queensland. His primary interests are survey research and the quantitative modelling of consumer attitudes/behaviour, especially in relation to substantive questions of marketing and public policy. He is also an instructor on the ACSPRI training programs in research methods/quantitative modelling for the social and behavioural sciences.

Location: Room 100

All RC33 members are welcome to attend

Thursday July 12, 2012

Location:
University of Sydney Law School (Camperdown)
New Law School Building (F10)
Eastern Avenue, Camperdown Campus
The University of Sydney NSW 2006

6:30PM - Departure 'Man O War' Steps East (leaving promptly at 6:30pm)

7:30PM - Dinner 'The Loft', Doltone House, Jones Bay Wharf

Friday July 13, 2012

Location:
University of Sydney Law School (Camperdown)
New Law School Building (F10)
Eastern Avenue, Camperdown Campus
The University of Sydney NSW 2006

Location: Room L.L 022

1. Ulrich Krieger and Julie Korbmacher: The Survey of Health, Ageing and Retirement in Europe: A primer
2. Matthias Ganninger: Sampling for SHARE: Challenges and possible Solutions
3. Marije Oudejans: Multi-country fieldwork management in SHARE
4. Julie Korbmacher: Enhancing SHARE survey data: administrative records and new biomarkers
5. Ulrich Krieger: The cost of survey response – Experimenting with monetary incentives

Location: Room L.L 100

Participants:
Robert Ackland, Australian National University
Maurice Vergeer, Radboud University
Mark S. Handcock, University of California, Los Angeles
Lexing Xie, Australian National University
Michael Jacobson, University of Sydney
Rob Procter, University of Manchester

Location: Room L.L105

Best practice is a phrase that is much used, rarely defined, and applied to a wide range of activities from specific survey process components through to the whole of a complex international organisation. This session will discuss best practice within the context of Fitness for Purpose with speakers and participants relating experiences and lessons learnt in implementing best practice from the perspective of user, design and process quality.

The panel will be led by Bill Blyth, Global Methods Director of TNS and Chair of ISO TC 225, the body overseeing the international survey quality standard. He will be joined by a panel that will include Beth-Ellen Pennell, Director of Survey Research Operations at the Survey Research Center, Institute for Social Research at the University of Michigan and Peter Lynn, Professor of Survey Methodology, University of Essex

Location: Room L.L 020

This roundtable session will focus on current issues and challenges in data archiving.

Panelists will include:
Steven McEachern, Deputy Director, Australian Data Archive
Peter Granda, ICPSR
Achim Wackerow, GESIS

Location: Room L.L 102

Facilitator: Trevor Webb, Food Standards Australia New Zealand

This roundtable session provides an informal opportunity to reflect on the papers presented over the previous days and the implications for how we do research related to food for policy and regulatory purposes. An initial summary of key themes emerging from the delivered papers will set the scene followed by open discussion.

Questions to be considered in the session could be:  What are the key challenges in measuring food choices and food behaviours going to be in the future? How might social sciences help to identify upcoming policy challenges? How do we need to strengthen/build our capacity for measuring food choices/behaviours? If we met again in 4 years what kinds of topics will we be researching and what methods will we be using?

Location: Room L.L 020

Facilitator: Samuel Spencer, Australian Bureau of Statistics

As a side event to RC33 the DDI Developers Group will be running a workshop for new and establish developers to promote the use of the Data Documentation Initiative metadata standards to help support social science research. As is the case with previous DDI Developers meetings there is no fixed agenda, instead participants are encouraged to bring demonstrations of their own work, talking points for discussion or real-world examples of DDI to help test against other's demonstrations. The goal of these meetings is to provide a casual environment to allow developers and users to discuss their strategies for implementing the DDI standards and give live demonstrations allowing fellow users of the standard to examine in-depth solutions to technical issues.

Note: Attendance to the DDI Developers Down Under meeting does not require registration for the main conference session. However, it is recommended you register for a free ticket at http://ddidownunder.eventbrite.com.au/ to ensure organisers are able to accurately track attendance.