ACSPRI Social Science Methodology Conference

Sunday December 7, 2014

Workshop presenters: Roy Lay-Yee and Barry Milne

Location: Woolley Lecture Theatre - S325

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. The workshop will be lecture-based. No pre-requisites or technical expertise are required. All welcome.

Workshop presenter: Steven McEachern

Location: Woolley Tutorial Room - N384

Workshop presenters: Aaron Martin and Ben Goldsmith

Location: Woolley Tutorial Room - N401

Many social scientists are turning to experiments as a way of overcoming the problems that plague studies based on observational data and, in turn, improving causal inference. This workshop is designed to provide an introduction to and discussion of various types of experimental research, and potential applications for Australian social science.

Workshop presenter: Adam Zammit

Location: Woolley Tutorial Room - S361

The first part of this workshop focuses on the use of Limesurvey, the free/open source web based survey tool. This is the tool that powers the ACSPRI Members Surveys service. Limesurvey allows for the creation of simple and complex web based questionnaires along with sample/list management.

The workshop concludes with an overview of the additional open source software made available by ACSPRI that allows for data collection in multiple modes, including paper based questionnaires and telephone interviews. These software integrate with the Limesurvey package allowing for a questionnaire to be defined once and then executed in multiple modes.

Bring along a laptop if you want a more hands-on experience.

Workshop presenter: Delwyn Goodrick

Location: Woolley Lecture Room - N497

Experimental design has been emphasised in research communities as the most rigorous design for causal analysis. There is growing interest about the capacity of qualitative research approaches to support causal analysis. The debates in the social research community reflect differing conceptions of causality underpinning approaches. Quantitative design and analysis procedures often focuses on what extent variance in one dimension causes variance in another. Qualitative design and analysis procedures are likely to be focused on understanding the role of one dimension in causing or influencing another dimension.

This workshop will provide an overview of the current perspectives on causal claims. A range of techniques that support causal analysis that draw on qualitative (small n, context-oriented) approaches. Qualitative comparative analysis, process tracing and applied thematic analysis will be discussed and reviewed. The comparative case study will be drawn on as an illustration of a promising design strategy for generating plausible causal claims.

The workshop will be of interest to conference participants that have some experience with qualitative research methods and are interested in robust approaches to qualitative data analysis.

Monday December 8, 2014

Survey Research in the 21st century: challenges, opportunities and open questions

Simon Jackman - Stanford University

Location: Footbridge Theatre

Survey research has occupied central stage in the social sciences for the best part of a century. Much has changed over that time: the populations surveys seek to study, the technology and methods available to survey researchers and the scientific ambitions and expectations of users of survey data. Drawing on my stewardship of the American National Election Studies — the longest running and best known research project in political science — I consider several related developments posing challenges and opportunities for survey researchers:

(1) declining response rates and the threat of non-response bias;
(2) “non-probability” sampling;
(3) respondents self-administering surveys via the Internet.

ANES uses a mix of in-person interviews (with samples generated by probability-based, multi-stage sampling designs, a method used since the project’s origins in 1952) and self-administered surveys with a variety of recruitment schemes. The cost of in-person interviewing is reaching unsustainable levels. But can long-standing, near-canonical, public-use surveys like ANES use non-probability samples (e.g., opt-in Internet panels) or self-administration? I review how ANES is wrestling the inevitable tradeoffs. We are increasingly attracted to a dual-
frame sampling design, recruiting a random sample from a mix of address-based sampling and person-level lists, with self-completion via the Internet.

This design is similar to the mail-out/mail-back, self-complete design used by the Australian Election Study, just one of number of points of connection with Australian survey research I will highlight.

---

Simon Jackman is Professor of Political Science and (by courtesy) of Statistics at Stanford University and a Visiting Professor at the United States Studies Centre at the University of Sydney. Jackman’s teaching and research centers on the application of statistical methods in the social sciences, with a particular focus on issues in democratic politics: public opinion, election campaigns, political participation, and electoral systems. Jackman’s research has appeared in the American Political Science Review, the American Journal of Political Science, the Journal of Politics, Political Analysis, the British Journal of Political Science, Electoral Studies and the Australian Journal of Political Science. A book length treatment of Jackman’s work on Bayesian statistical methods, Bayesian Analysis for the Social Sciences was published in 2009.

Jackman currently serves as one of the Principal Investigators of the American National Election Studies, the longest running and most authoritative survey-based study of political behavior and attitudes. Prior to his stewardship of ANES, Jackman directed a number of other large, on-line survey projects in the 2008 U.S. presidential election cycle. He is well known for his work on poll-averaging (combining polls over the course of an election campaign) and

Jackman studied for his doctorate at the University of Rochester (1989-1991) and Princeton University (1991-94), after graduating with Honours in Government from the University of Queensland. His first academic appointments were at the University of Chicago (1994-96) and the Australian National University (1996/97). Jackman has taught at Stanford since 1997. In 2004-05 Jackman served as President of the Society for Political Methodology, a 1,000 strong association of scholars with interests in methodological issues that arise in the scientific study of politics.

Location: The Refectory

Sponsored by The Social Research Centre

Tuesday December 9, 2014

Using solicited diaries in social research: Documenting the lived experience over time

Alex Broom - University of Queensland

Location: Footbridge Theatre

The use of solicited diaries as a means of collecting data is an established methodological tool. However, their use has not been widespread in the social sciences. Although diaries may lack the dialogical complexities and probing allowed in verbal communication, they also allow an examination of seemingly mundane day-to-day thoughts, processes and undulations. This design may be used very effectively to access the content of daily life for research participants and to transcend the potential artificiality and power dynamics of the face-to-face interview. A primary and significant benefit of personal diaries is the temporal nature of the insight they offer, allowing for flexibility and variation in the narratives presented.
 
In order to illustrate the potential of this design within the social sciences, I draw on the results of my recent study utilising solicited diaries with Australian women living with chronic back pain. The power of solicited diaries in understanding daily life and the lived experience was vividly evident within the research process. Utilising diary methods enabled the collection of richer, temporal and nuanced stories; more so than is possible in qualitative interviews. The diaries captured the inherent unpredictability of daily emotional fluctuations, the enmeshment of illness experience and life events, as well as providing a strongly participant-driven form of narrative construction. While there was some cost to removing the direction of the researcher, this was compensated by the authenticity, depth and scope of the solicited diary. I will conclude with discussion of the potential for wider utilisation of solicited diaries in social research.

---

Alex Broom is Associate Professor of Sociology, Australian Research Council Future Fellow and Head of Sociology at the School of Social Science, The University of Queensland, Australia. He specialises in the sociology of health, illness and wellbeing. Alex works regularly with a wide range of industry partners (e.g. hospitals, community organisations, professional organisations related to health and medicine) with a focus on improving people's experiences of illness and the delivery of healthcare. His program of research melds the conceptual richness of sociology with the value of applied, translational health research.

The Social Survey in Historical Perspective

Charlotte Greenhalgh - Monash University

Location: Footbridge Theatre

In the twentieth century a series of social surveys—distinguished by their first-person interviews and evidence-based recommendations—offered Australians a new form of knowledge about themselves. Far from being a simple discovery of fact, the act of social inquiry communicated a host of new ideas about the place of research participants within social hierarchy, state activity, and systems of knowledge. Importantly, social research conferred new status on the opinions and experiences of ordinary people, which were presented as scientific evidence that justified policy recommendations and academic publications alike.

Internationally, historians are familiar with the broad contours of the social scientific sea change in twentieth-century life, but not with its inner workings. When did social research become an expected part of modern life? What processes rearranged social hierarchies and professional practices in its image? And what transformations did social research produce on the ground, among research participants and consumers of social scientific and statistical knowledge? This paper’s reading of raw survey data—including interview transcripts, observations, and questionnaires—aims to recover the experience of participation in social research, to trace its consequences for national culture and politics, and to establish Australia’s role in the global development of social science.

---

Charlotte Greenhalgh is a Discovery Early Career Researcher at Monash University (2014-2017). She was previously the holder of a Commonwealth Scholarship at Oxford University (2009-2012) and a Kate Edgar postdoctoral award at the University of Auckland (2013). Charlotte’s doctoral research (Oxford, 2012) explored the emotional experience of ageing in mid-twentieth-century Britain. She has also published on young people, romance, and courtship in interwar New Zealand. Her current project examines the experience of participating in mid-twentieth-century social scientific research and its influence on modes of selfhood and social control in Britain, Australia, and New Zealand.

Wednesday December 10, 2014

The Social Scientist's Role in the Era of Big Data

Rob Ackland - Australian National University

Location: Footbridge Theatre

Social scientists are increasingly using large-scale datasets from the Web (e.g. Twitter, WWW hyperlinks, Facebook etc.) to seek answers to long-standing questions about social, economic and political behaviour. For example, social media data have been used to study social inequality, diurnal and seasonal mood changes and the spread of protest during the Arab Spring. This presentation aims to highlight the methodological challenges and opportunities of Big Data (in particular, social media data) in empirical social science research. There are two prevailing views on how Big Data will transform social science. One view is that theory and interpretation will become less necessary – data will “speak for themselves”. In this presentation, I argue for the counter view, namely that social science context, theory and models are required to identify meaningful correlations (and hopefully causality) in Big Data.

---

Robert Ackland is an Associate Professor in the Australian Demographic and Social Research Institute at the Australian National University and leads the Virtual Observatory for the Study of Online Networks (http://voson.anu.edu.au). His PhD was in economics, focusing on index number theory in the context of cross-country comparisons of income and inequality. Robert has been studying online social and organisational networks since 2002 and his research has been funded by five Australian Research Council grants. His research has appeared in journals such as the Review of Economics and Statistics, Social Networks, Computational Economics, Social Science Computer Review, and the Journal of Social Structure. VOSON was established in 2005, and aims to advance the social science of the Internet by conducting research, developing research tools, and providing research training. The VOSON software for hyperlink network construction and analysis has been publicly available since 2006 and has been used by over 1500 researchers worldwide. Robert established the Social Science of the Internet specialisation in the ANU's Master of Social Research in 2008, and his book Web Social Science: Concepts, Data and Tools for Social Scientists in the Digital Age (SAGE) was published in July 2013.

More information can be found here: https://researchers.anu.edu.au/researchers/ackland-rj

It's a wrap

Ali Dastmalchian - University of Victoria, Canada

Location: Footbridge Theatre

In this closing session Ali Dastmalchian will provide a commentary on the state of play of social research methods based on what he heard as he attended the plenaries and sessions throughout the conference. In particular, the focus will be on trends and emerging issues in this dynamic area.

---

Ali Dastmalchian is Professor of Organisational Analysis at The Peter B. Gustavson School of Business, University of Victoria. His main research interests include: organisational design, change and flexibility, organisational, human resource and industrial relations climate, cross-national and comparative research, and cross national leadership. He has published widely with articles in Academy of Management Executive, International Journal of Human Research Management, Human Relations, British Journal of Industrial Relations and Labor Relations Review.