1160.0 - ABS Confidentiality Series, Aug 2017  
Latest ISSUE Released at 11:30 AM (CANBERRA TIME) 23/08/2017  First Issue
   Page tools: Print Print Page Print all pages in this productPrint All RSS Feed RSS Bookmark and Share Search this Product

ABS MICRODATA: USES AND IMPACT ON RESEARCH QUALITY


This page contains the following:
ABS microdata products
The impact of data treatment on microdata
See also: Glossary of terms

ABS MICRODATA PRODUCTS

The Australian Bureau of Statistics (ABS) releases three broad microdata products that can be accessed for statistical and research purposes:

  • Basic Confidentialised Unit Record Files (CURFs) which can be loaded into a user’s own computing environment
  • Expanded CURFs, available through the Remote Access Data Laboratory (RADL) and the ABS DataLab
  • Detailed microdata files, available through the ABS DataLab.

Each product is designed to meet a different research requirement. For example, in Expanded CURFs the data reduction applied is less extensive than in Basic CURFs. A comparison of the three products is shown in Table 1.


TABLE 1: DIFFERENT ABS MICRODATA PRODUCTS
Microdata product
purpose
Access via
Confidentiality controls applied
Data treatment
Context controls

Basic CURFs
Basic utility
  • Surveys
  • Census samples
  • Simple modelling
  • Multivariate analysis
  • Exploratory analysis
  • ABS DataLab
  • Direct download
  • Direct identifiers removed
  • Data available at broad levels only (e.g. national)
  • Variables aggregated (e.g. 10 year age groupings)
  • At-risk records suppressed/removed
  • User registration required
Expanded CURFs
High utility
  • Surveys
  • Census samples
  • Complex modelling
  • Detailed analysis
  • Remote Access Data Laboratory (RADL)
  • ABS DataLab
  • Direct identifiers removed
  • Data available at finer levels (e.g. SA1 geography)
  • Limited aggregation applied (e.g. age in single years)
  • At-risk records suppressed/removed
  • RADL restricts the amount of information viewable by user and limits some system commands
  • ABS DataLab applies controls to the authorisation of people, to the projects undertaken and to the IT/physical access environment, and a clearance process is applied to the final output
Detailed microdata files
Very high utility
  • Surveys
  • Census datasets
  • Complex integrated administrative data
  • Complex modelling
  • Detailed analysis
  • ABS DataLab
  • Direct identifiers removed
  • At-risk records masked/suppressed/removed
  • ABS DataLab applies controls to the authorisation of people, to the projects undertaken and to the IT/physical access environment, and a clearance process is applied to the final output



THE IMPACT OF DATA TREATMENT ON ANALYSIS

Treating the data themselves may restrict the ability of a researcher to answer a particular question. For example, a major difference between Basic CURFs and Expanded CURFs is that data item categories in the former have been collapsed or aggregated to a greater degree, which reduces the level of detail available. In some instances it may therefore be more appropriate to use an Expanded CURF. Similarly, it may be better to conduct research (within the ABS DataLab) using detailed microdata files if the Expanded CURF does not contain enough detail to answer the researcher’s question.
There are, however, situations where data treatment may not adversely affect the quality of the data or the reliability of the research. For example, a 2010 study used data from the ABS’s Survey of Mental Health (2007) to compare results attained from using the Expanded CURF with results attained from using the untreated main-unit-record file (MURF). As Table 2 shows, the results were almost identical.

TABLE 2: HAZARD RATIOS* FOR SMOKING CESSATION AND INCIDENT OF ANXIETY DISORDERS
Disorder
Hazard ratio (from CURF)
Hazard ratio (from MURF)
Standard error of
hazard ratio
Difference between hazard ratios, as proportion of standard error

No lifetime mental disorder
1.00
1.00
(reference category)
.
Anxiety disorder (type):
Panic disorder
0.59678
0.59615
0.10335
0.0102
Agoraphobia
0.44936
0.44936
0.08420
0
Social phobia
0.55374
0.55374
0.07973
0
Generalised anxiety disorder
0.33631
0.33631
0.07846
0
Obsessive-compulsive disorder
0.47782
0.47782
0.10674
0
Post-traumatic stress disorder
0.63221
0.63221
0.07865
0
.
Anxiety disorder (severity):
Mild
0.74901
0.74882
0.11495
0.00218
Moderate
0.58942
0.58920
0.08275
0.00447
Severe
0.39196
0.39202
0.06016
0.00249

* Hazard ratios compare the incidence of an event in one group to another group over time


Source: Lawrence, D., Considine, J., Mitrou, F. & Zubrick, S.R. (2010) ‘Anxiety disorders and cigarette smoking: Results from the Australian Survey of Mental Health and Wellbeing’, Australian and New Zealand Journal of Psychiatry. Vol. 44, pp. 521-528.