Regional Report 1996-1997 Asia/Pacific

Report of the Australian Regional Secretary

by Vance Merrill
Social Science Data Archives
Australian National University
Canberra ACT
May 1997

During the past year, IASSIST gained an additional Australian member (Len Smith, National Centre for Epidemiology & Population Health, the Australian National University), yet most activity related to Association aims continued to be carried out by the Social Science Data Archives (SSDA) and ACSPRI (Australian Consortium for Social and Political Research Incorporated). Whereas the SSDA acquires, preserves and distributes social science data to Australian researchers; ACSPRI encourages and supports activities which facilitate and enhance use of these data. Membership of ACSPRI has grown by 5 in the past two years, and the Consortium currently has 45 member institutions, including all 36 of Australia’s publicly funded universities and 9 Commonweath agencies.

Since the last report, the SSDA, in revamping its web pages (, implemented networked access to its documentation and data by placing a number (currently 16) of its User’s Guides online and by developing its Online Analysis System in the Social Sciences (OASSIS), enabling users to browse, search and/or download (in RTF) codebooks before selecting (and recoding) variables for simple (frequencies & crosstabs) analysis online. For more extensive analysis, data sets may be ordered online by, and FTP’d to, users. The ACSPRI Newsletter/SSDA News and the current Course Booklets for both the ACSPRI Summer and Winter Programs in Social Research Methods and Research Technology may be browsed/downloaded as well.

During the past two years, ACSPRI has conducted two more annual Summer Programs at the ANU, and two additional Winter Programs in Adelaide and Brisbane (in 1995 and 1996, respectively). In spite of substantial Federal budget cuts to academia and the public service, all four Programs were successful in attracting more participants then had the immediately preceding Programs. The recently concluded 13th Summer Program, for example, attracted 282 participants to twenty courses, and was, in terms of the total number of courses taken, the largest to date.

However, the previous 1996 Summer Program was noteworthy for having attracted research librarians from 8 Australian universities, the Commonwealth Parliamentary Library, and the Australian Bureau of Statistics (ABS) to a special workshop, conducted as part of the Program, on ‘Mainstreaming Social Science Data: Providing Data Services in the Library’. Taught by North American IASSIST members Diane Geraci, Chuck Humphrey and Jim Jacobs (aka the ‘data amigos’), the five-day workshop, while not immediately productive in terms of recruiting new Australian members to the Association, succeeded in familiarising participants with data library ‘best practices’ as they might apply to, and/or be modified by, local circumstances.

None of the participating librarians had previous ‘hands-on’ computer readable data processing experience, but they rated the workshop highly, as did the ‘data amigos’ who, on departing Australia, wrote: “We had a great time! And no complaints (well, except for the urinals and the fact that you can’t get a decent cookie). Really enjoyed meeting your colleagues…. " (from: Amigos, Data, “It’s time we face reality, my friends…. We’re not exactly rocket scientists.", Thank You Card, Canberra: Jan. 1996.).

SSDA/ACSPRI continue to work to facilitate academic user access to ABS unit record data, though an ‘understanding’ reached in early 1995 between ACSPRI/SSDA and the ABS for distribution of the Bureau’s existing and proposed confidentialised unit record files, or CURFs, did not achieve formal agreement. That arrangement would have improved academic access to ABS unit record data by halving the cost of purchases by academic institutions from $5000 to $2500; currently, however, only the Time Use Survey 1992 and Family Survey 1993 are being distributed under the above pricing arrangement.

The period of uncertainty regarding arrangements for the dissemination of CURFs to academia coincided with the appointment in 1995 of the current Australian Statistician. His subsequent internal review of the Bureau’s structure and policies resulted in a substantial reorganisation of the senior executive levels and a (seemingly) more restrictive interpretation of the Census and Statistics Act (and the most recent Ministerial Determination) that guide release of ‘unidentifiable unit records’.

According to the C&S Act, the ABS cannot release data ‘in a manner that is likely to enable the identification of a particular person or organisation’; the Determination requires recipients to sign an undertaking that they must not attempt to identify individuals or organisations, that the information will be used only for statistical purposes and that it will not be disclosed to other individuals or organisations. Moreover, to quote the Statistician: “If I consider it necessary, I can impose further conditions…This Determination is an enabling Determination; it does not compel me to release unit record data. Before I authorise the release of microdata I need to be satisfied that ABS guarantees in respect of confidentiality are not only met, but are seen to be met.” And, since “this is a very sensitive area” his decisions, which are informed by advice from the ABS’s Microdata Review Panel, “are made on a case by case basis” (from: McLennan, W., “The Product of The Australian Bureau of Statistics”, Austral. J. Statist. 38(1), 1996, 1-14).

In the event, the ABS is now charging (academic and non-academic) purchasers $7500 for each CURF created after 1 July 1996; no new CURFS are available for secondary distribution by the SSDA; and no CURFs can be distributed outside of Australia. It was in this context that a feature article, based on materials generously supplied by Wendy Watkins and Chuck Humphrey, on the Canadian Data Liberation Initiative (March ‘96 ACSPRI Newsletter) generated enthusiasm among data users Down Under, and sparked a cautious response from the ABS (see Sept. ‘96 ACSPRI Newsletter), which is now discussing ways and means.