Social Statistics

Aims:

The principal aims of a Social Statistics Section in the SSAI are to:

  1. Advance research in social statistics within Australia in areas which involve use of methods of statistical enquiry, and use of statistical data and the development of statistical measurement.
  2. Foster links between the statistical and social science communities within Australia and New Zealand, and with similar communities internationally.
  3. Promote and support the development and use of statistical techniques that are valuable in the analysis of social data.
  4. Raise awareness of the archives of social data and activities of large-scale social science data collection being undertaken within Australia and their potential use in improving knowledge of social outcomes and in providing evidence towards the formulation of social policies.
  5. Promote the role of statistics in the social science disciplines within universities and government.

There is a need within Australia to encourage and support research in the development and application of statistical techniques to social data that has been collected through surveys, administrative records and other methods. Australia is undergoing major social and economic changes and in response to this change governments have been focusing on innovative and interdisciplinary “social science” that uses advanced methodologies to generate new knowledge and a rigorous evidence base for policy development. Consequently there is a need to develop better ways of managing and analysing data to address these changes and to inform social policy.

There are many challenging statistical research questions in social science research that require the development of new or extended statistical methodology. Nevertheless, the statistical knowledge applied to social research problems in Australia is often limited. Areas that require promotion for the analysis of social data include:

  • methods for longitudinal data (often ordinal or nominal),
  • the application of multilevel models to clustered data,
  • hierarchical models for spatially integrated social data,
  • the application of event-history analysis,
  • dealing with missing data, and
  • the application of Bayesian analysis using Markov chain Monte Carlo simulation techniques, to name a few.

A Section in Social Statistics in the SSAI is an ideal way to begin promoting the role of statistics in social science disciplines. Both the American Statistical Association and the Royal Statistical Society have Sections in Social Statistics to facilitate collaborative activities and training. Conversely, the American Sociological Association has a Section in Quantitative Methodology to promote the development and use of statistical methodology in social research.

Intended Membership:

  • Statisticians working and teaching in universities, government and industry in the areas of applied social research across a range of social science disciplines
  • Statistical advisors and consultants
  • Social scientists in government, universities and research centres using statistical methods for advancing social research
  • Postgraduate students in areas of social research and applied statistics

Proposed Activities:

  • Establish links with various national social research centres and government departments to promote the importance of statistical methodology in social research. The current Chair has existing links with UQ ISSR and other social research centres through this association. She also has links with the Department of Family and Community Services and other government departments through research collaboration.
  • Organise and maintain a website through the SSAI and an email list of social researchers and statisticians who would like to keep up-to-date with the activities of the Section.
  • Organise a regular seminar series and meeting to bring together statisticians and social researchers.
  • Organise and facilitate short courses on various types of statistical methodology for social data in collaboration with social research centres and government departments e.g. methods for analysis of longitudinal categorical data.
  • Organise and chair sessions in statistical methodology at conferences held by social research centres such as MIAESR, ACSPRI and UQ ISSR.
  • Organise and chair sessions in statistical methodology for social research at Australian and New Zealand statistical conferences and encourage social researchers to present

To join the Social Statistics Mailing List please click here.

Chair:

Professor Michele Haynes
Program Leader, Research Methods and Social Statistics
Institute for Social Science Research
The University of Queensland
Brisbane, Queensland, 4072

Phone: +61 7 3346 9690
Email: m.haynes@uq.edu.au

Assistant Chair:

Ms. Louise Marquart
Mr. Arturo Martinez Jr.
Institute for Social Science Research
The University of Queensland
Brisbane, Queensland, 4072

Email:  l.marquart@uq.edu.au / a.martinez2@uq.edu.au

 

 

News:

The findings of two cutting-edge research on complementary aspects of social statistics will be presented in a webinar to be held in the morning of 20th November 2014, UK time. This is part of the activities of the Royal Statistical Society’s Journal Club.

 “
The item count method for sensitive survey questions: Modelling criminal behaviour
The item count method is a way of asking sensitive survey questions which protects the anonymity of the respondents by randomization before the interview. It can be used to estimate the probability of sensitive behaviour and to model how it depends on explanatory variables. The results of the author’s analysis of criminal behaviour highlight the fact that careful design of the questions is crucial for the success of the item count method.
Speakers: Jouni Kuha and Jonathan Jackson

Which method predicts recidivism best? A comparison of statistical, machine learning and data mining prediction models
Risk assessment instruments are widely used in criminal justice settings all over the world. However, in recent times, different approaches to prediction have been developed. This paper investigates whether modern techniques in data mining and machine learning provide an improvement in predictive performance over classical statistical methods such as logistic regression and linear discriminant analysis. Using data from criminal conviction histories of offenders, these models are compared. Results indicate that in these data, classical methods tend to do equally well as or better than their modern counterparts.
Speakers: Nikolaj Tollenaar and Peter van der Heijden

Chair: Professor Chris Skinner, professor of statistics at the London School of Economics & Political Science.