6102.0.55.001 - Labour Statistics: Concepts, Sources and Methods, 2006  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 15/12/2005   
   Page tools: Print Print Page Print all pages in this productPrint All  
Contents >> Appendix

INTERNATIONAL GUIDELINES ON DISSEMINATION AND ABS PRACTICE

As statistics generally have assumed a more important role in the day to day activities of modern economies and societies, a number of international agencies including the International Monetary Fund and the ILO have put forward a range of proposals and guidelines for the dissemination of data; the methodology of its collection and compilation, and evaluation as to its accuracy; relevance to the phenomena measured; and quality of the output.

In particular, the ILO at its 1998 ICLS endorsed a set of 20 guidelines concerning dissemination practices for labour statistics. How these compare with current ABS release practices is outlined in the following section.


2.1 LABOUR STATISTICS DISSEMINATION - ICLS GUIDELINES AND ABS PRACTICE


ICLS GUIDELINES

ABS PRACTICE

.

.

ACCESS

1Ready access should be provided to data and metadata and they should be actively disseminated and publicised.

Subject to national data protection restrictions, micro-data that protect the confidentiality of respondents should be made available - at least for research purposes.
The ABS public good policy (outlined above) ensures ready access to the data from its statistical collections. Unpublished data are also generally available on request.

Metadata are disseminated alongside the data they relate to in explanatory notes, glossaries, commentary, and other material associated with each statistical collection.

Micro-data, also referred to as unit record data, are available from a number of statistical collections. The release of unit record data is subject to strict release provisions to protect confidentiality. Unit record data are made available to a number of Australian universities for academic and teaching purposes.
.
2A variety of statistical products should be derived from each data set, using appropriate dissemination media in each case.
.
A wide range of products is used to disseminate ABS statistics including publications, information papers, articles and data reports. Information is available in printed and electronic form.
3The professionalism of statistical agency staff should encompass the skills and techniques needed to design tables and charts, to communicate information effectively to users; also presentational and media skills.
.
The ABS employs professional staff who are trained in a variety of disciplines, and possess the technical skills necessary to design, collect, process and disseminate information to users.
4The initial release of the main labour aggregates into the public domain should be free of charge; an explanation of any charging policies that are followed by the statistical agency for additional outputs should be publicly available.
.
Main labour aggregates are released into the public domain free of charge. Charging policies and practices for more detailed data are publicly available.
5The professional staff responsible for labour data should ensure their names and/or workplace telephone numbers are published with all statistical outputs; other forms of direct contact with users such as user groups and 'help lines' should be developed by the statistical agency.Contact names and details of staff responsible for statistical collections are published with all outputs from those collections.

The ABS provides a range of services to address users' general information needs including national statistical service, dial-a-statistic, library, Internet, and consultancy services.

The ABS consults regularly with users to obtain feedback on directions for its statistical collections. In the field of labour statistics, an advisory group composed of key users of labour data has been formed as a forum to seek users' views on a range of labour statistics issues.
.
6Statistical agencies should regard the provision of data and metadata to international organisations as equivalent in importance to the supply of data to home customers; international organisations should adopt dissemination guidelines themselves.
.
The ABS makes data freely available to a number of international organisations.


.



.


INTEGRITY

7The terms and conditions under which statistics are produced and released, including labour statistics, should be a matter of public record.The terms and conditions under which statistics are produced are outlined in the Commonwealth Census and Statistics Act 1905 which describes the statistical information to be collected, the administration of statistical activities, obligations to answer questions, the publication of statistics, the release of other information, secrecy provisions, and the fines and penalties that can be imposed for the failure to comply with requests for information, or for the furnishing of false or misleading statements.
.
8Those agencies that have technical responsibility and prepare analytical commentary on the data should have responsibility for releasing data.
.
The ABS has responsibility for the collection, compilation, analyses and release of data.
9There should be prior announcement of the date of initial release of labour statistics including, where possible, the exact time. The earlier the advance notice the better, even if dates are issued initially on a provisional basis. When release deadlines are not met, the reason should be made publicly available.Release dates are set in advance for ABS publications and other standard products.

The release dates for main economic indicators are set and published 12 months in advance. The release date for the next publication is advised in each main economic indicator release. Labour-related main economic indicators include:
  • Labour Force Survey (Preliminary);
  • Wage Cost Index; and
  • Job Vacancies.

Other regular monthly and quarterly release dates are set and published six months in advance. The release date for the next publication is advised in each release. Labour-related monthly and quarterly releases include:
  • Industrial Disputes;
  • Average Weekly Earnings; and
  • Employment and Earnings.

The month of release is set and published for annual and irregular releases twelve months in advance. The day of release is published three months prior to the release. Annual and irregular labour-related releases include:
  • Employee Earnings and Hours;
  • Labour Costs;
  • labour-related supplementary surveys; and
  • labour-related Special Social Surveys.
10Data should be released the same day and at precisely the same time of day to all parties. If special privileges are given to journalists, this should be under 'lock-up' conditions.
.
All ABS data are embargoed prior to 11.30am on the day of their release. They are then simultaneously released in both electronic and printed copy. Pre-releases and lockups are severely restricted. If circumstances require either lockup or pre-release they are limited to a short period of time prior to general release.
11If demands for prior access to the data within governments cannot be resisted, the number of persons with advance access should be kept to an absolute minimum and a list of these persons should be maintained; the period of notice they are given should be kept as short as possible.
.
Prior access is only given for main economic indicators and then only to Ministers responsible for government policy covered by that field of statistics, and to their departments. The pre-release is limited to a short period of time prior to embargo time. Special conditions apply to ensure that the pre-release information is only used to prepare Ministerial comment at or shortly after the official release time.
12Ministerial commentaries and statements made at the time the statistics are released must be clearly distinguished from those of the statisticians.
.
Ministerial comment is only made after the official release of ABS data, and is clearly distinguishable from ABS analysis.


.



.



DATA
13Where there are significant shortfalls in the coverage of national employment and unemployment totals, users should be made aware of this and reminded every time the data are released. Providing subtotals sufficiently reliable can be produced for the purpose to be served, their availability should be well publicised.
.
The ABS attempts to provide as many dissections of the data as accuracy will permit. Extensive use is made of Explanatory Notes, caveats and information papers to inform the public of any shortcomings in the data.
14Similarly, users should be made well aware of the reference period of the data.
.
The reference period associated with each collection is highlighted in publication titles, commentary and all associated output.
15If countries have the resources and there is evidence of user demand, the main aggregates should be released at least on a quarterly basis. Publication of information necessary for detailed structural analysis of the labour market should occur at least annually.
.
Estimates of the labour force are released monthly, while industry and occupation splits are released quarterly.

A further range of labour data is produced at quarterly, annual and less frequent intervals.
16Labour statistics should be released as soon as possible after the data have been assembled and analysed. When the source is a household or establishment survey conducted monthly or quarterly, data for the main aggregates should normally be available within a quarter of the end of the reference period to which they refer. Ideally, annual survey data should be released, at least on a preliminary basis, within a half year of the reference period.ABS policy is to release monthly main economic indicators within the month following the end of the reference period; other monthly publications are released within 6 weeks of the reference period.

Quarterly main economic indicators should be released within two months of the reference quarter; other quarterly publications should be released within the quarter following the reference quarter.

Labour force supplementary surveys should be released within six months of the reference period.

Annual publications should be released within one year of the reference period.

For irregular collections the first major release should be within 12 months of the reference period.

Also see point 9 above.
.


.



.



QUALITY
17Regularly updated documentation on metadata - the definitions, methodology, sources, sampling error and other quality indicators, the questionnaires, forms, etc., used in preparing the statistics - should be made publicly available, including the degree of alignment with international recommendations.
.
Metadata are disseminated along with the data to which they relate, in the explanatory notes, glossaries, commentary, and other explanatory material associated with statistical outputs.

Major changes to forms, questionnaires or methodology are announced in special information papers.
18Where there are two or more sources of labour data, reconciliation or comparisons between them should be published regularly. Statistical frameworks and accounting schemes that support statistical cross-checks should also be developed.
.
Reconciliations or comparisons and reasons for differences are published from time to time, usually in explanatory notes within publications, or in special information papers.
19Users should be given adequate advanced warning of revisions and their implementation should be guided by a code of practice.
.
Users are given adequate advance warning of major revisions, usually via an information paper prepared and published well in advance of implementation.
20The statistical agency should make estimates for missing periods whenever collection or collation of data is interrupted. Similarly, the effects of discontinuities should be estimated.In cases where there are large discontinuities, the effects of these discontinuities are estimated using the best methodology available.






Previous PageNext Page