Stephen Horn (stephen.horn@fahcsia.gov.au) and John Preston (john.preston@abs.gov.au)
This Section focuses on the methodology behind official statistics production storage dissemination and deployment and use. It invites participation from members interested or involved in theory, design, collection, management, storage and presentation of population, enterprise and process data in public policy sphere. It has had a close association in the past with Australian Bureau of Statistics through its survey methodology units; now represents the broadening sector of statisticians engaged outside the ABS at different levels of government. It has links with the international community through IAOS and IASS – the arms of ISI dealing with official and survey statistics respectively, and with more specialised bodies with scope in these broad fields.
The Section has organised one-day workshops at the last few Australian Statistics conferences on topics of current interest – notably panel surveys, and missing data. In the non-ASC years, half-day workshops have been organised, either in Wollongong or Canberra or both. Included have been sessions on data linkage, confidentiality and imputation, attracting participants from across Australia and New Zealand.
It would like to extend such events to cover emerging issues in the region and looks forward to cooperation with other Sections and Branches in the society on these and other possible projects. The Section aims to represent the professional interests of all major employers of government statisticians, and consultants working principally in the public sphere. It remains independent of employers, and respects the normal protocols concerning sharing and ownership of public assets, or information accessed under privilege in the course of employment.
On behalf of the society the Section is exploring cooperative arrangements with allied professional organisations.
Surveys and Management Section
Events
2012 Panel Survey Methods Workshop
The 3rd Panel Survey Methods Workshop will be held 4-5 July 2012 at the University of Melbourne, Australia. The aim of the workshop is to foster discussion and initiate methodological research specific to the collection of panel survey data. The format of the meeting will be informal, designed to encourage interaction and collaboration. Presentations will be limited to a 10 minute summary of key findings or ideas, in order to allow generous time for discussion. All participants will have had a chance to study materials in advance, as presenters will be asked to circulate a handout before the workshop. The handout should provide background information upon which the presentation will be built. One person from each panel study represented at the workshop will be invited to also provide a 2-page summary of the recent and planned key methodological developments and these will be circulated prior to the workshop. The meeting will end with a brainstorming session, to identify a research agenda of the key issues.
The focus is primarily on surveys that involve collecting data from subjects on multiple occasions (panel surveys). Note that the workshop is concerned with the data collection methods used by such surveys, not with the substantive findings. Topics of interest include, but are not limited to, the following:
·Measurement error, including panel conditioning, seam effects, recall error, and dependent interviewing;
·Mixed mode data collection, including effects on cost, measurement error, attrition, and logistics;
·Attrition and non-response, including effects of survey design features, assessment of bias, methods to minimise effects, and adjustment methods (weighting and imputation);
·Role of interviewers in longitudinal surveys;
·Paradata, including collection and use in longitudinal surveys;
·Sampling issues, including refreshment sampling and following rules;
·Innovations in longitudinal surveys, including biomarkers, online panels, matrix sampling, remote access, linking data sources, consent issues, and experimental designs; and
·Use of administrative data in longitudinal surveys.
Papers may provide a review of research and practice in a particular methodological area, present the findings of new methodological research or outline planned research for which the presenter would like feedback. Submissions will be screened by the organising committee, who will select papers that address relevant topics and collectively provide a balanced program.
Registration for the workshop is free of charge. The organizers have secured funding to provide subsidies of $2000 (AUD) to 20 participants to assist with travel costs. While we would like to encourage active participation, presentation of a paper is not a requirement for participation in the workshop. Please note that places will be limited to about 50 people.