Hilda News - 2002
2013 2012 2011 2010 2009 2008 2007 2006 2005 2004 2003 2002 2001
18 December 2002
Progress Report, Wave 2, HILDA Survey
Fieldwork for Wave 2 of the HILDA Survey has been progressing well. As of about a week ago we had identified a sample of 8236 households (wave 1 responding households plus splits minus known dead households). Of these, interviews had been achieved at 7053, and at the large majority (90%) interviews have been completed with ALL in-scope adults.
So far only 56 households have been deemed untraceable, though there are a further 173 households that are still in the tracking process.
While the main fieldwork components are now 95% complete, there is a third phase that will commence in February and finish in March where we hope to track down some of the non-contacts and convert some of the less strident refusers. Merry Xmas
16 October 2002
As you should all be aware, the confidentialised data file for the HILDA Survey Wave 1 was released yesterday. As part of that release, we have published our first Annual Report, which summarises our activity since the HILDA Project commenced.
In addition, FaCS released three papers which examine various aspects of the wave 1 data. These papers are as follows:
"Fertility Pathways in Australia: Relationships, Opportunities, Work and Parenting", by Kathleen Fisher, FaCS.
"Youth in Focus: Interaction Between Educational Participation, Education Attainment and Labour Force Activity", by Val Pawagi, FaCS.
"Labour Force Patterns and Self-perceived Health Status Among Older Australians: Implications for Healthy Ageing", by Wiliwan Kanjanapan, Dept of Health and Ageing.
All of these papers plus the latest HILDA Annual Report can now be downloaded from the HILDA website.
28 October 2002
HILDA Conference - Call for papers
Most of you should be aware that the 2003 HILDA Conference about wave 1 data is being organised for March 13 next year. While much of the time has been allocated to invited speakers, we have room for at least three papers from other users. We are thus calling for abstracts for possible papers. Authors are free to choose the topic they wish to consider, but use of wave 1 data from the HILDA Survey must, of course, be central to the paper. Abstracts should be sent to Professor Mark Wooden by end of November.
Expenses associated with paper presenters attending the conference including airfares and accommodation will be reimbursed (but limited to one presenter per paper).
16 October 2002
As you should all be aware, the confidentialised data file for the HILDA Survey Wave 1 was released yesterday. As part of that release, we have published our first Annual Report, which summarises our activity since the HILDA Project commenced.
In addition, FaCS released three papers which examine various aspects of the wave 1 data. These papers are as follows:
"Fertility Pathways in Australia: Relationships, Opportunities, Work and Parenting", by Kathleen Fisher, FaCS.
"Youth in Focus: Interaction Between Educational Participation, Education Attainment and Labour Force Activity", by Val Pawagi, FaCS.
"Labour Force Patterns and Self-perceived Health Status Among Older Australians: Implications for Healthy Ageing", by Wiliwan Kanjanapan, Dept of Health and Ageing.
All of these papers plus the latest HILDA Annual Report can be downloaded from the HILDA web site.
10 October 2002
With the release date of the Wave 1 data almost here, I can now inform you that the Coding Framework for that data set is now available as a Technical Paper on the HILDA web site. Note that this paper will be included as part of the documentation accompanying the unit record data.
2 October 2002
If you are interested in obtaining the unit record files of the HILDA Survey Wave 1 data (available on CD-Rom), the order form can now be downloaded from the HILDA web site.
All persons wishing to obtain the data must complete the order form and sign the accompanying Deed of License (which can also be downloaded from the same web page), and then send original copies of both documents to:
HILDA Survey, HILDA Administrative Assistant
Melbourne Institute of Applied Economic & Social Research
6th floor, Economics and Commerce Building
University of Melbourne VIC 3000
AUSTRALIA
Note that there is a charge of $75 for the data for users within Australia. Also note that Licenses are only issued to individuals and not organisations.
23 September 2002
HILDA Survey Progress
Wave 1
- As reported previously, the unit-record public-release file for the first wave of the HILDA Survey will be launched in Canberra on October 15. Approved users will be able to obtain the data (on CD-Rom) from that date. The License which will govern use of these data is still with the FaCS legal team, but once it has been finalised, it, together with an application form, will be posted on the website. We will advise when the application form is available.
- An "Annual Report" will be released at the same time. All subscribers to this list will automatically be sent a copy in the mail (assuming address details supplied were, and are still, correct of course).
- The Coding Framework will form part of the documentation that accompanies the data. However, it is also being released as part of our Technical Paper series and hence will be able to be downloaded from the HILDA web site (by the end of September).
- The process of data checking revealed a few problem cases, including cases that appeared to have been double-entered. On further checking, we actually found some respondents where the responses were overwritten by other previously recorded cases. The total number of responding cases is thus actually greater than previously advised. The total number of individuals completing the Person Questionnaire is thus 13969.
- The latest issue of the Australian Economic Review (September 2002) contains a short article about Wave 1 outcomes. Apart from summarising key features of the survey design, it also provides a summary of sample characteristics. The full reference for that article is:
Mark Wooden, Simon Freidin and Nicole Watson (2002), 'The Household, Income and Labour Dynamics in Australia (HILDA) Survey: Wave 1', Australian Economic Review, vol. 35, no. 3 (September), pp. 339-348.
- Interviewing for Wave 2 is now underway in all States.
- Funding for Wave 3 is close to finalisation. At this stage it appears extremely likely that we will continue to use face-to-face interviewing as the principal mode of data collection. This involves a substantial increase in Commonwealth funding over the initial budget. Nevertheless, the additional funding does not go so far that it can cover the additional cost of a special module (as was done with Wealth in Wave 2). Individual interview time in Wave 3 will thus be about 10 minutes less than in Wave 2.
5 August 2002
HILDA Survey Progress
Wave 1
- The unit record public-release file for the first wave of the HILDA Survey will be launched in Canberra on October 15. Approved users will be able to obtain the data (on CD-Rom) from that date. Details of the user access policy together with a data request form will be made available on the HILDA website in the near future. We have set a price of $75 to cover handling costs.
- The HILDA data launch is being combined with the Melbourne Institute Public Economic Forum. It is being held at Parliament House, Canberra, on October 15 starting from 11.30am. Anyone wishing to attend should contact:
Before August 19 - Ms Rachel Derham, 03 8344 5325
After August 19 - Ms Penny Hope, 03 8344 7885. - I recently conducted a seminar for the Victorian Government on features of the Wave 1 data. If other groups are interested in having me (or one of our staff) speak, give us a call.
- Interviewing for Wave 2 commences later this month.
- All survey instruments for Wave 2 can now be accessed from the HILDA web site.
- While early days, a conference discussing preliminary results from early uses of the Wave 1 data, has been proposed for March 13 in 2003. The venue is yet to be confirmed but is likely to be in either Melbourne or Canberra. I am especially hearing from anyone who has plans on using the data as soon as it's released. Likely presenters include Dan Hamermesh (University of Texas), and Stephen Jenkins (University of Essex), as well as staff of the Melbourne Institute, FaCS, and the Department of Health and Ageing.
23 May 2002
Final totals for the Wave 1 questionnaires are as follows:
In-scope households: 11,693
Interviewed households: 7,682 (6,872 full-response from all members of the household aged 15 or more years)
Personal interviews: 13,965
Self completion interviews which can be matched to persons interviewed: 13,051 (93.4% of persons interviewed)
Data cleaning is in full swing, with numerous checks being made within and between the survey questionnaires. In particular an effort has been made to overcome some of the limitations of the optical character recognition employed for the self-completion questionnaires, with an examination of the scans for every multi-marked data item. This has resulted in the reduction of the total number of multi-marked items by two-thirds. In addition there have been lookups of unusual patterns of data (often caused by strike-throughs), large amounts of missing data, and unusual numeric responses.
The creation of survey weights is well underway.
Planning for the derived variables to be appended to the wave 1 dataset for the October release is ongoing.
31 January 2002
The fieldwork for Wave 1 of the HILDA Survey is now complete, and all indicators are that the survey was a great success.
The final sample, after adjusting for out-of-scope dwellings and multiple households, numbered 11,693. A complete set of interviews was achieved with all household members at 6,871 of these households, with interviews being completed with at least one household member at a further 809 households. The achieved sample thus consists of 7,680 households (though these numbers are still subject to some final edit checks) and represents a household response rate of 66%.
We have yet to determine how many individual interviews this has generated, but we expect somewhere close to 14,000 cases.
The incidence of missing data items also appears to be relatively low.
Preliminary analysis of response rates by cluster indicates that response was highly variable across region, being lowest in the Northern Territory and in Sydney, and highest in rural SA and WA.
There also appears to be an income effect, with response rates about 5 to 6 percentage points lower in high income areas than elsewhere. Response rates in low income areas, however, were not, other things equal, any different than in middle-income areas.
Areas with high concentrations of non-English speaking households were also more likely to have low response rates.
There do not appear to be any significant variations in response rates with person age, average household size, household composition or labour force status.
Final response rates, together with further details about the fieldwork outcomes will be made available on the HILDA web site in due course.