Microdata: National Survey of Mental Health and Wellbeing, Basic and Expanded CURF

Provides information about the data content of the 2007 National Survey of Mental Health and Wellbeing

Introduction

Overview

This technical manual provides information about the release of microdata from the 2007 National Survey of Mental Health and Wellbeing (SMHWB), including background to the survey, using the CURF microdata, reliability of estimates, contents of the 2007 CURFs and the conditions of release. Microdata from the survey are released with the approval of the Australian Statistician and are available in the form of two Confidentialised Unit Record Files (CURFs):

  • a Basic CURF, available on CD-ROM or through the ABS Remote Access Data Laboratory (RADL); and
  • an Expanded CURF, only accessible through the RADL.

The Expanded CURF contains more detailed data for some variables than the Basic CURF, as well as some additional variables. For the contents of each CURF refer to the data item list, which is available on the ABS website in spreadsheet format.

The RADL is an online database query system that supports access to the CURFs. Microdata are held on a server at the ABS and accessed by submitting programs to interrogate and analyse data, as well as access the results. More information about accessing the CURFs and using the RADL is available from the ABS website (see Services/ CURF Microdata/ Accessing CURF Microdata).

Information to aid interpretation of the CURFs is provided in this paper and in the National Survey of Mental Health and Wellbeing: Users' Guide (cat. no. 4327.0), which will be referred to throughout this paper as the Users' Guide.

About the survey

The National Survey of Mental Health and Wellbeing (SMHWB) was conducted by the Australian Bureau of Statistics (ABS) from August to December 2007. The survey collected information from approximately 8,800 Australians aged 16-85 years. The survey provides information on the prevalence of selected lifetime and 12-month mental disorders, by the major disorder groups:

  • Anxiety disorders (eg Social Phobia);
  • Affective disorders (eg Depression); and
  • Substance Use disorders (eg Alcohol Harmful Use).

The survey also provides information on the level of impairment, health services used for mental health problems, physical conditions, social networks and caregiving, as well as demographic and socio-economic characteristics.

This is the second national mental health and wellbeing survey conducted by the ABS, with the previous survey conducted in 1997. Funding for this survey was provided by the Australian Government Department of Health and Ageing (DoHA). The survey is based on a widely-used international survey instrument and is therefore designed to provide data for international comparisons, rather than comparisons with the 1997 survey.

Detailed information on the survey development, concepts and methodology is provided in the Users' Guide.

Using the CURF microdata

About the microdata

The survey microdata are released under the provisions of the Census and Statistics Act 1905. This Act allows for the release of data in the form of unit records where the information is not likely to enable the identification of a particular person or organisation. Accordingly, there are no names or addresses of survey respondents on the CURFs, and other steps have been taken to protect the confidentiality of respondents. These include:

  • reducing the level of detail shown for some items (eg geography, industry and occupation items);
  • ranging, collapsing or top- or bottom-coding some data items (eg age first time consulted with a health professional); and
  • changing characteristics for some records (eg masking information by changing demographic details for a small number of records).

To further assist in protecting the confidentiality of unit record data, all dollar values have been perturbed (Expanded CURF) or are presented in deciles only (Basic and Expanded CURFs). That is, each value has been adjusted up or down by a small, random amount. Also, for each of these items, each value above or below a certain cut-off limit has been set to the mean of all the values above or below the cut-off limit.

Steps to confidentialise the data available on the CURFs are taken in such a way as to maximise the usefulness of the content while maintaining the confidentiality of respondents to ABS statistical collections. As a result, it may not be possible to exactly reconcile all the statistics produced from the CURFs with published statistics.

Intending purchasers should ensure that the data they require, at the level of detail they require, are available on the CURFs. Data collected in the survey but not contained on the CURFs may be available on request in tabulated form.

Data items

A detailed list of the data items and categories for the Basic and Expanded CURFs has been released on the ABS website in spreadsheet format. The majority of data items on the CURFs are the same as collected and output on the main survey file. In a small number of instances it was necessary to remove items or to reduce the level of information available on the CURFs for confidentiality reasons. Where information on the CURFs differs to the main survey file, this is noted as 'Not on the CURF' or the details of the CURF output items are included on the data item list. Refer to the Basic/Expanded CURF SAS name and CURF item (including response categories) on the data items list for more information.

Additionally, some of the data items included on the CURFs, and the categories within the data items, differ between the Basic and Expanded CURFs. The Expanded CURF contains more variables than the Basic CURF, as well as more detailed response categories for selected variables. The differences between the CURFs are highlighted in the data item list.

Key differences between the CURFs

Demographic items

Most of the differences between the Basic and Expanded CURFs relate to the amount of detail available for the demographic items. The following table provides an overview of the key differences.

Key differences between the CURFs, by demographic items
ItemBasic CURFExpanded CURF
Household level
State/Territory-NSW, Qld, Vic, SA, WA, Tas, NT, ACT
Number of persons in household6 categories8 categories
Number of adults in household4 categories5 categories
Number of children aged 0-17 years in household4 categories5 categories
Family composition of household8 categories11 categories
Index of disadvantage (CD level)DecilesDeciles
Capital city/Rest of State-Capital city, Balance of State
ASGC remoteness area category-Major cities, Inner regional, Other areas
Gross weekly cash income of household in deciles-Deciles
Gross weekly cash income of household

-

Continuous income $ (perturbed)

Gross weekly equivalised cash income of household-Continuous income $ (perturbed)
Person level
Social marital status-Not applicable, Married, Not married
Country of birth of person/father/mother3 categories20 categories
Main language other than English spoken at home (ASCL)-9 categories
Level of highest non-school qualification9 categories10 categories
Highest level of post-school educational attainment7 categories9 categories
Occupation10 categories53 categories
Industry21 categories70 categories
Personal gross weekly cash income-Continuous income $ (perturbed)
Personal gross weekly cash income in deciles-Deciles

Note: - not available

Service use items

There are also differences between the Basic and Expanded CURFs for items related to service use in the Health Service Utilisation module, including:

  • the number of times a person was admitted to hospital for physical health problems in the 12 months prior to interview;
  • the total number of nights a person was admitted to hospital for physical health problems in the 12 months prior to interview;
  • the number of consultations with a [health professional] for physical or mental health in the 12 months prior to interview;
  • the number of consultations with a [health professional] for mental health in the 12 months prior to interview; and
  • the age first time consulted with a [health professional] about mental health problems.

This information is collected separately for the following different types of health professionals consulted:

  • General Practitioners (GPs);
  • Psychiatrists;
  • Psychologists;
  • Mental health nurses;
  • Other professionals (specialist mental health);
  • Specialist doctors/surgeons;
  • Other professionals (general); and
  • Alternative therapists.

The Expanded CURF data items generally contain more detail than was able to be provided on the Basic CURF. For more information on the data items available and the applicable response categories refer to the data item list.

CURF file structure

The Basic and Expanded CURFs each contain a set of two files with confidentialised records. These files are in a hierarchical relationship and provide records at the household and person levels. Details of the information contained in the household and person level files for the Basic and Expanded CURFs are provided in the data item list. There are 8,841 records on each file level for both of the CURFs.

Use of weights

On each level of the CURF, every record contains a 'weight'. The weight indicates how many population units are represented by the sample unit. The CURFs contain records which can be adjusted (weighted) to infer results for the total in-scope population in Australia. As the survey was conducted on a sample of private dwellings in Australia, it is important to take account of the method of sample selection when deriving estimates from the CURFs. If the chance of selection is not accounted for, by use of appropriate weights, the results will be biased.

Each household and person record contains a weight. The identifiers are:

  • Person weight: MHSFINWT, and
  • Household weight: HHFINWT.

The household weight can be found on the household level, while the person weight can be found on the person level. This weight indicates how many population units are represented by the sample unit.

Where estimates are derived from the CURF, it is essential that they are calculated using the correct weight for that level of the file, and not just counting the number of records in each category. If person or household weights were to be ignored when analysing the data to draw inferences about the population, then no account would be taken of a person’s or household's chance of selection, or of different response rates across population groups, and the resulting estimates may be seriously biased. The application of weights ensures that estimates conform to an independently estimated distribution of the population by age and other characteristics, rather than to the distributions within the sample itself.

It should be noted that as a result of some of the changes made to protect confidentiality on the CURF, estimates of benchmarked items produced from the CURF may not equal benchmarked values. For more information about these differences see 'About the Microdata'.

Replicate weights

Replicate weights have been included on the CURFs to enable the calculation of sampling error for each estimate produced. Sampling error arises because the estimates are based on a sample of units and so will differ from estimates that would have been produced if all units in the population had been included in the survey. The CURFs contain:

  • 60 person replicate weights (WPM0101 to WPM0160); and
  • 60 household replicate weights (WHM0101 to WHM0160).

For more information on sampling error see 'Reliability of estimates'.

Record identifiers

There are several identifiers for each record on each of the levels:

  • Each household has a unique random identifier (ABSHID). This identifier appears on the household level, and is repeated on the person level for each record relating to that household.
  • Each person within a household is numbered sequentially. 'Person number' (ABSPID) appears on the person level. The combination of household identifier and person number uniquely identifies a person.

At higher levels, identifiers for lower levels are set to zero. For example, on the household level the identifier for person number is set to zero. No family numbers or income units have been output for this survey.

Deriving items at a higher level

The information provided on the files has already been derived at the relevant level. For example, the number of adults in a household is already provided on the household level and does not need to be manually derived from data on the person level.

Copying data to a lower level

There may be instances when a data item is not contained on the level of the file required by users. For example, geographic information is not included on the person level because it is a household level characteristic. To copy data from the household level to the person level:

  • sort household and person level by household. All records for a particular person must be sorted together. This can be done by sorting on the identifiers for household (ABSHID); and
  • merge the records in household and person level by household. The records must be matched using the household identifier (ABSHID).

These resulting dataset will contain all the household level information (eg geography) attached to the person level data (eg age and sex).

Continuous data items

When analysing continuous items at the person and household levels, it is necessary to exclude the special codes. The special codes are used for responses that do not represent the data being collected (eg 'don't know'). The codes vary, but will generally be 0, 96, 97, 98, 99 or variations of these. For example, 'Length of time was homeless in days' (DAYSHMLS) has reserved values of:

  • 999,996 for Not stated;
  • 999,998 for Not known; and
  • 999,999 for Refusal.

The data item list provides the special codes for continuous items. Care should be taken to exclude these codes when categorising higher values for ranges, and when calculating means, medians and other summary statistics.

Income items

The person level records contain the following information on income:

Basic CURF

  • source(s) of personal cash income (SINCHHM); and
  • main source of personal cash income (MSRCINCC).

Expanded CURF

  • personal gross weekly cash income in deciles (INCDECPN);
  • personal gross weekly cash income in continuous perturbed dollar amounts (WINCCEX);
  • source(s) of personal cash income (SINCHHM); and
  • main source of personal cash income (MSRCINCC).

The household level records contain information derived from the income of all persons in the household, aged 15 years and over:

Basic CURF

  • gross weekly equivalised cash income of households in deciles (INCDECEH); and
  • sources of cash income of households (INCSRCH).

Expanded CURF

  • gross weekly cash income of households in deciles (INCWKHHD);
  • gross weekly cash income of households in continuous perturbed dollar amounts (WINCPHX);
  • gross weekly equivalised cash income of households in deciles (INCDECEH);
  • gross weekly equivalised cash income of households in continuous perturbed dollar amounts (INCWKHEX); and
  • sources of cash income of households (INCSRCH).

When analysing income at the person and household level, it is necessary to exclude the special codes (eg 'not stated'). The data item list provides the relevant values.

Income not reported

If more than one contributing income item at the person level has a value of 'Not known', then totals derived from these items, such as 'Gross weekly personal income' are also set to 'Not known', as it was not possible to derive an accurate total. Similarly, if one or more of the contributing person records in a household has a value of 'Not known', then household income and derived income deciles are set to 'Not known/no income reported'. Care should be taken to exclude these codes when categorising higher income values, and when calculating means, medians and other summary statistics.

Calculating income deciles

Deciles are groupings that result from ranking either all households or all persons in the population in ascending order according to some characteristic, such as income, and then dividing the population into 10 equal groups, each comprising 10% of the estimated population. The first decile contains the bottom 10%, the second decile contains the next 10% and so on.

To assist in the use and interpretation of income deciles at the person or household level, it is necessary to exclude the reserved values of:

  • 00 for Not applicable;
  • 98 for Not stated; and
  • 99 for Not known.

Refer to 'Income not reported' for more information on the treatment of 'Not known or not stated' values.

The following table provides the decile ranges for gross weekly cash income of persons (INCDECPN) and of households (INCWKHHD):

Gross weekly cash income of persons and households, by decile ranges
 Personal incomeHousehold income
Decile$$
1st< 152< 399
2nd153 - 229399 - 576
3rd230 - 299577 - 843
4th300 - 432844 -1 084
5th433 - 5991 085 -1 360
6th600 - 7601 361 -1 642
7th761 - 9581 643 -2 017
8th959 -1 1882 018 -2 539
9th1 189 -1 6392 540 -3 321
10th>1 640>3 321

Gross weekly equivalised income of households

Equivalised household income provides an indication of the economic resources available to each member of a household. It can be used for comparing the situation of individuals, as well as comparing the situation of households. Equivalence scales are used to make adjustments to the actual incomes of households so that households of differing sizes and compositions are able to be compared. The gross weekly equivalised income of households is expressed in:

  • deciles (Basic and Expanded CURFs); and
  • continuous perturbed dollar amounts (Expanded CURF only).

Equivalised income is calculated by deriving an equivalence factor according to the chosen equivalence scale, and then dividing income by the factor. The equivalence factor, derived using the 'modified OECD' equivalence scale, is determined by allocating the following points to each person in a household:

  • the first adult in the household is given a weight of 1 point;
  • each additional person 15 years or older is allocated 0.5 points; and
  • each child under 15 years is allocated 0.3 points.

Equivalised household income is derived by dividing the total household income by a factor equal to the sum of the equivalence points allocated to the household members. The equivalised income of a lone person household is the same as its unequivalised income. The equivalised income of a household comprising more than one person lies between the total value and the per capita value of its unequivalised income.

When unequivalised household income is negative, such as when a loss is reported for an individual's unincorporated business or other investment income, and this loss is greater than any positive income from any other source, then equivalised household income is set to zero.

When analysing the gross weekly equivalised income of households, it is necessary to exclude the special codes (eg 'not stated'). The data item list provides the relevant values.

The following table provides the decile ranges for gross weekly equivalised income of households (INCDECEH):

Gross weekly equivalised income of households, by decile ranges
 FromTo
Decile$$
1st <277
2nd278359
3rd360479
4th480613
5th614755
6th756897
7th8981 064
8th1 0651 303
9th1 3041 713
10th>1 713 

Multiple response items

There are a number of data items on the Basic and Expanded CURFs that contain multiple responses. This means that the person being interviewed was able to select one or more response categories for these items. Multiple response items are indicated on the data item list.

On the CURF, each response category for the multiple response questions is treated as a separate data item. Each data item therefore has a response of either:

  • Not applicable; or
  • Yes.

A 'Not applicable' response has a code of '0' indicating that the response category does not apply for the respondent. A 'Yes' response has a code greater than '0' indicating a positive response for that category.

An example of a multiple response item is the question on the 'Type of health problem causing embarrassment related to social or performance situation' (HPSOCFR), which has eight response categories. From these categories eight separate data items have been produced - HPSOCFRA, HPSOCFRB, HPSOCFRC...HPSOCFRH.

Type of health problem causing embarrassment related to social or performance situation
Response categoryData itemData item response
Mental health problemHPSOCFRANot applicable (0), Yes (1)
Alcohol or drug problemHPSOCFRBNot applicable (0), Yes (2)
Speech, vision or hearing problemHPSOCFRCNot applicable (0), Yes (3)
Movement or coordination problemHPSOCFRDNot applicable (0), Yes (4)
Facial/body disfigurement or weight/body image problemHPSOCFRENot applicable (0), Yes (5)
Bad odour or sweatingHPSOCFRFNot applicable (0), Yes (6)
PregnancyHPSOCFRGNot applicable (0), Yes (7)
Other physical health problemHPSOCFRHNot applicable (0), Yes (8)
Not knownHPSOCFRINot applicable (0), Yes (98)
RefusalHPSOCFRJNot applicable (0), Yes (99)
Not applicableHPSOCFRKNot applicable (0), Yes (97)

The special codes for 'Not applicable', 'Not known' and 'Refusal' are provided in the data item list (eg values of 97, 98 and 99). For derived items, these codes each have their own separate data item category. Using the example above, these special codes would be HPSOCFRI, HPSOCFRJ and HPSOCFRK. For non-derived items (eg D23A) the special codes will be placed within the first data item created (eg D23AA).

Subsequent questions relying on multi-response items

Multi-response items in the Medications, Substance Use and Mini-Mental State Examination (MMSE) modules used a different system to code the multiple response items. This meant that depending on the responses provided to an initial question, the responses for subsequent questions were allocated to a certain position.

An example of these linked questions in the Medications module is MEDCD, where respondents could provide up to five responses (ie medications) to the question on 'Mental health medications used in past 2 weeks'. The subsequent A to E responses provided for the following items relate back to the number of medications provided in the initial question:

  • Length of time taking medication for mental health (ME6A--ME6E); and
  • Whether usually took medication according to the recommended dose in past two weeks (ME7A--ME7E).

For example, if Citalopram was reported at MEDCDA then the length of time the respondent had been taking Citalopram would be reported at ME6A, and whether the Citalopram was taken according to recommendation would be reported at ME7A. Similarly, responses at ME6B and ME7B would be in relation to the medication reported at MEDCDB.

Multi-response data items
Data item labelData item name
Mini-Mental State Examination
Whether remembered the words Ball/Car/ Man (immediate recall)MM3
Whether could subtract 7sMM4
Whether remembered the words Ball/Car/ Man (delayed recall)MM6
Substance Use disorders
Whether able to follow instructions - take paper, fold in half, place on floorMM13
Whether used drugs of marijuana, stimulants, sedatives or opioids caused problems with family, friends or othersSU65A1A
Whether continued using drugs of marijuana, stimulants, sedatives or opioids caused problems with peopleSU65B1A
Whether continued using drugs of marijuana, stimulants, sedatives or opioids jeopardised safetySU65C1A
Whether continued using drugs of marijuana, stimulants, sedatives or opioids resulted in problems with policeSU65D1A
Whether used drugs of marijuana, stimulants, sedatives or opioids interfered with workSU65_1A
Whether had strong desire using marijuana, stimulants, sedatives or opioids that one could not resist themSU721_A
Whether needed to use more marijuana, stimulants, sedatives or opioids than used to, in order to get highSU72A1A
Whether experienced symptoms when stopped using marijuana, stimulants, sedatives or opioidsSU72B1A
Whether used marijuana, stimulants, sedatives or opioids from having problemsSU72C1A
Whether used marijuana, stimulants, sedatives or opioids though promised not toSU72D1A
Whether used marijuana, stimulants, sedatives or opioids more frequently than intendedSU72E1A
Whether not able to stop using marijuana, stimulants, sedatives or opioidsSU72F1A
Whether spent much time using marijuana, stimulants, sedatives or opioids that had little time for anything elseSU72G1A
Whether reduced important activities because of using marijuana, stimulants, sedatives or opioidsSU72H1A
Whether continued using marijuana, stimulants, sedatives or opioids knowing one had a problemSU72I1A
How much has physical health been harmed by use of marijuana, stimulants, sedatives or opioids during the past 12 monthsSU86A_1
How much has family been hurt by use of marijuana, stimulants, sedatives or opioids during the past 12 monthsSU86B_1
How much have done impulsive things by use of marijuana, stimulants, sedatives or opioids during the past 12 monthsSU86C_1
How much has failed to do what was expected by use of marijuana, stimulants, sedatives or opioids during the past 12 monthsSU86D_1
Substance Use section - Interviewer checkpoint (see SU86 series)SU86G
Number of days unable to work in the past 12 months - Drug UseSU86G1
Medications
Mental health medications used in past 2 weeksMEDCD
Length of time taking medication for mental healthME6
Whether usually took medication according to the recommended dose, in past 2 weeksME7

Geographic characteristics

To enable CURF users greater flexibility in their analyses, one Socio-Economic Index for Area (SEIFA) has been included on both the Basic and Expanded CURFs, as well as several sub-state geography items. Cross-tabulations by several of these items simultaneously produce cells relating to some small geographic regions. Tables showing multiple data items, cross-tabulated by SEIFA and sub-state geography are not permitted due to the detailed information about small geographical regions that could be presented. However, simple cross-tabulations of population counts by SEIFA or sub-state geographic data items may be useful for clients in order to determine which geography item to include in their primary analysis, and such output is permitted. Note: the state/territory data item is only available on the Expanded CURF.

More information on geographic characteristics is contained in Chapter 9 (Population characteristics) of the Users' Guide. For more details on SEIFA see the Information Paper: An introduction to Socio-Economic Indexes for Areas (SEIFA) (cat. no. 2039.0), available from the ABS website.

Comparison with the 1997 CURF

The 2007 survey was designed to provide national estimates that can be compared internationally, rather than to provide comparisons with the 1997 survey. Due to differences in how the data were collected, care should be exercised when comparing data items from the 1997 survey with the 2007 survey. Particular attention should be given to the definition of the data item, the population, and the reference period that applies (eg 12-month versus lifetime). Differences between the two surveys are too substantial to list individually in this paper, but include changes to:

  • questions and topics;
  • concepts;
  • survey methodology;
  • classifications; and
  • measurements.

Detailed information on the differences between the two surveys is provided in the National Survey of Mental Health and Wellbeing: Users' Guide, 2007. More information about the 1997 survey is available from the National Survey of Mental Health and Wellbeing of Adults: Users' Guide, 1997 (cat. no. 4327.0).

Reliability of estimates

Sample survey errors

Two types of error are possible in estimates based on a sample survey:

  • sampling error; and
  • non-sampling error.

Sampling error occurs because only a small proportion of the total population is used to produce estimates that represent the whole population. Sampling error can be reliably measured as it is calculated based on the scientific methods used to design surveys.

Non-sampling error may occur in any data collection, whether it is based on a sample or a full-count (eg Census). Non-sampling error may occur at any stage throughout the survey process, examples include: non-response by selected persons, questions being misunderstood, responses being incorrectly recorded, and errors in coding or processing the survey data. More detailed information on sample survey errors, including sampling error, non-sampling error and response rates is provided in Chapter 10 (Interpretation of results) of the Users' Guide.

Sampling error

Sampling error is the expected difference that could occur between the published estimates, derived from repeated random samples of persons, and the value that would have been produced if all persons in scope of the survey had been included. The magnitude of the sampling error associated with an estimate depends on the sample design, sample size and population variability.

Measures of sampling error

A measure of the sampling error for a given estimate is provided by the Standard Error (SE), which is the extent to which an estimate might have varied by chance because only a sample of persons was obtained.

Another measure is the Relative Standard Error (RSE), which is the SE expressed as a percentage of the estimate. This measure provides an indication of the percentage errors likely to have occurred due to sampling.

Standard errors of proportions and percentages

Proportions and percentages formed from the ratio of two estimates are subject to sampling errors. The size of the error depends on the accuracy of both the numerator and the denominator. The RSEs of proportions and percentages for the National Survey of Mental Health and Wellbeing: Summary of Results, 2007 (cat. no. 4326.0) were provided in spreadsheet format as an attachment to the publication. The RSEs were calculated using the full delete-a-group jackknife technique, which is described in the following segment.

Replicate weights and directly calculated standard errors

The SEs on estimates from this survey were obtained through the delete-a-group jackknife variance technique. In this technique, the full sample is repeatedly subsampled by successively dropping households from different groups of clusters of households and then the remaining records are reweighted to the survey benchmark population. Through this technique, the effect of the complex survey design and estimation methodology on the accuracy of the survey estimates is stored in the replicate weights. For the 2007 SMHWB, this process was repeated 60 times to produce 60 replicate weights for each sample unit. The distribution of the 60 replicate estimates based on the full sample estimate is then used to directly calculate the standard error for each full sample estimate.

Replicate weights enable variances of estimates to be calculated relatively simply. They also enable unit records analyses such as chi-square and logistic regression to be conducted, which take into account the sample design. Replicate weights for any variable of interest can be calculated from the 60 replicate groups, giving 60 replicate estimates. The distribution of this set of replicate estimates, in conjunction with the full sample estimate (based on the general weight) is then used to approximate the variance of the full sample.

The person level and household level records on the CURFs contain 60 replicate weights. The Standard Error (SE) for each estimate produced from the CURFs can be calculated using the replicate weights provided. When calculating SEs it is important to select the replicate weights which are most appropriate for the analysis being undertaken. For more information see 'Use of weights'.

The formula for calculating the Standard Error (SE) and Relative Standard Error (RSE) of an estimate using this technique is shown below.

\(SE(y) = \sqrt{((59/60) \sum_g(y(_g)-y)^2)}\)

where
\(_g = (1,...,60(the\space number\space of\space replicaweights))_i \)
\(y(_g) = estimate\space from\space u\space singreplicaweighting;\space and \)
\(y = estimate\space from\space u\space singfullpersonweight.\)

\(The\space RSE(y)={\frac{SE(y)}{y} \times 100}\)

This method can also be used when modelling relationships from unit record data, regardless of the modelling technique used. In modelling, the full sample would be used to estimate the parameter being studied, such as a regression co-efficient, the 60 replicate groups used to provide 60 replicate estimates of the survey parameter. The variance of the estimate of the parameter from the full sample is then approximated, as above, by the variability of the replicate estimates. For more information on the replicate weights technique refer to Appendix 2 of the Users' Guide.

Use of the delete-a-group jackknife technique for complex estimates, such as regression parameters from a statistical model, is not straightforward and may not be appropriate. The technique described does not apply to investigations where survey weights are not used, such as unweighted statistical modelling. More information on the delete-a-group jackknife technique is provided in the Research Paper: Weighting and Standard Error Estimation for ABS Household Surveys (Methodology Advisory Committee) (cat. no. 1352.0.55.029).

CURF users should be aware that estimates produced from the CURFs may differ from those in the published data due to actions taken to preserve confidentiality.

Non-sampling error

Efforts were made to minimise non-sampling error by careful design and testing of questionnaires, intensive training of interviewers, and extensive editing and quality control procedures at all stages of data processing. However, errors can be made in giving and recording information during an interview. These types of inaccuracies are referred to as non-sampling errors and include errors in the survey scope, response errors, processing errors and bias due to non-response.

Response rate

As the response rate for this survey was lower than expected (60%), extensive non-response analyses were undertaken to assess the reliability of the survey estimates. As a result, adjustments were made to the weighting strategy. As non-response can vary across population characteristics, as well as across data items, users should exercise caution. Detailed information on survey methodology and non-response is provided in the Users' Guide.

Contents of the CURFs

Basic CURF files

The Basic CURF is available either on CD-ROM or through the RADL. The available files formats are listed in the following table.

Basic CURF, file formats available on CD-ROM or through RADL
File levelSAS for WindowsSPSS for WindowsSTATA
HouseholdMHW07BH.SAS7BDATMHW07BH.SAVMHW07BH.DTA
PersonMHW07BP.SAS7BDATMHW07BP.SAVMHW07BP.DTA

Basic CURF on CD-ROM

In addition to the above-mentioned files, the CD-ROM contains raw confidentialised survey data in a comma separated value (CSV) file format:

  • MHW07B.CSV - the raw data for the Basic CURF; and
  • MHW07B.SAS - a SAS program which can be used to translate data from the CSV format to SAS. It also creates the formats library and attributes the formats to relevant variables. This file is for use with other analysis packages and describes the CSV file data. Relevant changes will need to be made to reflect the load statement for your analysis package.

Expanded CURF files

The Expanded CURF contains more detailed data than the Basic CURF and is only available via the RADL. The available files formats are listed in the following table.

Expanded CURF, file formats available through RADL
File levelSAS for WindowsSPSS for WindowsSTATA
HouseholdMHW07EH.SAS7BDATMHW07EH.SAVMHW07EH.DTA
PersonMHW07EP.SAS7BDATMHW07EP.SAVMHW07EP.DTA

Supporting documentation

CURF information

The following files are provided with the CURF/s:

  • FORMATS.SAS7BCAT - a SAS library containing formats.
  • README.TXT - a text file describing the contents of the CURFs.

Documentation of the data in plain text format is also provided. The following files contain data item code values and category labels with weighted and unweighted frequencies of each value.

Frequencies, documentation of the data in plain text format
File levelBasic CURFExpanded CURF
HouseholdFREQUENCIES_MHW07BH.TXTFREQUENCIES_MHW07EH.TXT
PersonFREQUENCIES_MHW07BP.TXTFREQUENCIES_MHW07EP.TXT

General information

The following files are provided with the CURF/s:

  • RESPONSIBLE ACCESS TO ABS CURFs TRAINING MANUAL_MAR05.PDF - this explains the role and obligations of people using confidentialised data;
  • ABS CONDITIONS OF SALE.PDF - this describes the ABS conditions of sale;
  • COPYRITE1.BAT - this describes the copyright obligations for people using ABS data and products; and
  • IMPORTANT INFORMATION FOR CURF USERS_300903.PDF - this directs people to the ABS website for additional and updated information.

The following related documentation is available from the ABS website:

  • Managing ABS Confidentialised Unit Record Files (CURFs): A step by step guide, Feb 2009 (cat. no. 1406.0.55.004);
  • ABS Remote Access Data Laboratory (RADL): User Guide, Mar 2006 (cat. no. 1406.0.55.002); and
  • Guide to citing ABS sources - this provides samples of citations for ABS data and products, including CURFs (see Help/ How do I cite ABS sources?).

Survey products

The following information is available from the ABS website:

Conditions of release

Release of the CURFs

The Basic and Expanded CURFs have been released in accordance with a Ministerial Determination (Clause 7, Statutory Rules 1983, No. 19) in pursuance of Section 13 of the Census and Statistics Act 1905. As required by the Determination, the CURFs have been designed so that the information on the files is not likely to enable the identification of a particular person or organisation to which it relates.

CURF user undertaking

The Australian Statistician's approval is required for each CURF release. Additionally, the ABS requires all organisations, and individuals within organisations, who purchase or are seeking to use a CURF to sign an undertaking to abide by the legislative restrictions on use before access to a CURF will be granted. The undertaking requires that people will:

  • use the information only for the statistical purposes specified in the Schedule to the Undertaking;
  • not attempt to identify particular persons or organisations;
  • not disclose, either directly or indirectly, the information to any other person or organisation other than members of their organisation who have been approved by the ABS to have individual access to the information;
  • not attempt to match, with or without using identifiers, the information with any other list of persons or organisations;
  • comply with any other direction or requirement specified in the 'Responsible access to ABS CURFs training manual'; and
  • not attempt to access the information after the term of their authorisation expires, or after their authorisation is rescinded by the organisation which provided it, or after they cease to be a member of that organisation.

Use of data for statistical purposes

Use of the data for statistical purposes means the CURF data is used to produce information of a statistical nature. Examples of statistical purposes are:

  • manipulation of the data to produce means, correlations or other descriptive or summary measures;
  • estimation of population characteristics;
  • use of data as input to mathematical models or for other types of analysis (e.g. factor analysis); and
  • providing graphical or pictorial representations of the characteristics of the population or subsets of the population.

All CURF users are required to read and abide by the 'Responsible access to ABS CURFs training manual' available from the ABS website (see Services/ CURF Microdata/ Accessing CURF Microdata). Use of the data for unauthorised purposes may render the purchaser liable to severe penalties. Advice on the propriety of any intended use of the data is available from the Microdata Access Strategies Section via microdata.access@abs.gov.au.

Conditions of sale

All ABS products and services are provided subject to the ABS Conditions of Sale. A copy of the conditions is available from the ABS website (see About Us/ ABS Conditions of Sale). Queries should be referred to Intermediary Management via intermediary.management@abs.gov.au.

While care is taken in handling each CURF on CD-ROM, deterioration may occur between the time of copying and receipt of the file. Accordingly, if the CD-ROM is unreadable on receipt and this is reported to the ABS within 30 days of receipt, it will be replaced free of charge.

CURF prices

From 1 January 2009 the recommended retail prices for access to ABS CURFs are:

  • $1,430 (including GST) per CURF access type (ie $1,430 to access the Basic CURF on CD-ROM and/or RADL, or $1,430 to access the Expanded CURF on RADL); or
  • $2,140 (including GST) for access to both the Basic and Expanded CURFs. Clients must request access to both the Basic and Expanded CURFs in one single application.

For more information on pricing and payment refer to the ABS website (see Services/ CURF Microdata).

Accessing the CURFs

Due to the level of detail provided, the Expanded CURF is only available via the ABS Remote Access Data Laboratory (RADL). The Basic CURF is available through both CD-ROM and RADL. The CURF is not available on CD-ROM to overseas customers.

Australian universities

The CURFs can be accessed by universities participating in the ABS/Universities Australia Agreement for research and teaching purposes. University staff and students should refer to the ABS website (see Services/ Services for Universities).

Other clients

Other prospective clients should contact the ABS Microdata Access Strategies Section:

Phone: (02) 6252 7714
Email: microdata.access@abs.gov.au
Mail: Locked Bag 10, Belconnen ACT 2617

Data downloads

Data files

Previous releases

 TableBuilder data seriesMicrodataDownloadDataLab
Mental Health and Wellbeing of Adults, 1997 Basic microdata 

Quality declaration

Institutional environment

Confidentialised Unit Record Files (CURFs) are released in accordance with the conditions specified in the Statistics Determination section of the Census and Statistics Act 1905. This ensures that confidentiality is maintained whilst enabling micro level data to be released. More information on the confidentiality practices associated with CURFs is available through the Microdata Entry Page.

For information on the institutional environment of the Australian Bureau of Statistics (ABS), including the legislative obligations of the ABS, financing and governance arrangements, and mechanisms for scrutiny of ABS operations, see ABS Institutional Environment.

Relevance

Microdata from the 2007 National Survey of Mental Health and Wellbeing (SMHWB) is available in the form of a Basic CURF and an Expanded CURF. A detailed list of the data items and categories for the CURFs has been released on the ABS website in spreadsheet format as an attachment to the Technical Manual: National Survey of Mental Health and Wellbeing, Confidentialised Unit Record Files (cat. no. 4329.0). Key differences between the data items for the two CURFs are also highlighted in the Technical Manual.

The survey provides information about the prevalence of selected mental disorders in the Australian adult population, the level of impairment associated with these disorders, physical conditions, and the use of health services, such as consultations with health practitioners or visits to hospital. The survey also provides information on the strength of social networks, caring responsibilities and a range of socioeconomic and demographic characteristics.

Timeliness

The survey was conducted from August to December 2007 and the CURFs released approximately fourteen months after enumeration. Other survey products include:

This is the second survey of this type conducted by the ABS, with the previous survey conducted in 1997. A CURF, 'Mental Health and Wellbeing of Adults, Australia, 1997', was released from the 1997 survey. More information is available via the 'List of available CURFs'. Publications from the previous survey are also available from the ABS website.

Accuracy

The CURFs contains household and person level data (unit records), and finer levels of detail for data items than what is otherwise published. For more information on the level of detail available from the CURFs refer to the data item list.

The survey was designed primarily to provide estimates at the national level, ie Australia. Due to the higher than expected non-response rate, the RSEs for all estimates are somewhat larger than originally designed. Broad estimates may be available at the state level for the larger states, eg New South Wales. However, users should exercise caution when using estimates at this level due to high sampling errors. Note: the state/territory data item is only available on the Expanded CURF.

Due to a lower than expected response rate (60%) extensive non-response analyses were undertaken to assess the reliability of the survey estimates. As a result, adjustments were made to the weighting strategy. As non-response can vary across population characteristics, as well as across data items, users should exercise caution. Detailed information on survey methodology and non-response is provided in the National Survey of Mental Health and Wellbeing: Users' Guide, 2007 (cat. no. 4327.0).

Steps to confidentialise the data available on the CURF/s are taken in such a way as to maximise the usefulness of the content while maintaining the confidentiality of respondents to ABS statistical collections. As a result, it may not be possible to exactly reconcile all the statistics produced from the CURF/s with published statistics.

Coherence

This is the second survey of this type conducted by the ABS, with the previous survey conducted in 1997. The 2007 survey was designed to provide data that are internationally comparable, rather than to provide comparisons with the 1997 survey. Additionally, there have been a number of changes to the scope, design, collection, methodology and content. Therefore, care should be exercised when comparing this survey to the previous survey. Detailed information on comparisons between the two surveys is provided in the National Survey of Mental Health and Wellbeing: Users' Guide, 2007 (cat. no. 4327.0).

Interpretability

To assist interpretation and understanding of the CURF microdata, these key sources should be consulted:

The Users' Guide includes information on the survey objectives, design, methodology, content, data quality, interpretation and comparability with the 1997 survey. The Technical Manual provides details on the content of the CURF files and a data item list, released as a spreadsheet attachment to the publication.

Additionally, the National Survey of Mental Health and Wellbeing: Summary of Results, 2007 (cat. no. 4326.0) provides a summary of the main findings from the survey, 15 tables with footnoted data and the relative standard errors for these tables.

Accessiblity

CURF microdata are not available to the public without special access being granted. All CURF users are required to read and abide by the 'Responsible Access to ABS Confidentialised Unit Record Files (CURFs) Training Manual'. Details on how to apply for access to CURF microdata can be completed and submitted for approval by following the steps listed in the How to apply for Microdata web page, available from the ABS website. A full list of available CURFs can be viewed via the Expected and available Microdata.

The Basic CURF can be accessed on CD-ROM and through the Remote Access Data Laboratory (RADL). The Expanded CURF can only be accessed through the RADL. More detail regarding the types and modes of access to CURFs is available from the Microdata Entry Page.

For queries about accessing the CURFs please contact the ABS Microdata Access Strategies Section:

Phone: (02) 6252 7714
Email: microdata.access@abs.gov.au

Abbreviations

Show all

The following symbols and abbreviations are used in this publication:

ABSAustralian Bureau of Statistics
CD-ROMcompact disc read-only memory
CSVcomma separated value
CURFconfidentialised unit record file
DoHAAustralian Government Department of Health and Ageing
GSTgoods and services tax
RADLRemote Access Data Laboratory
RSErelative standard error
SEstandard error
SEIFASocio-Economic Indexes for Areas
SMHWBNational Survey of Mental Health and Wellbeing

Previous catalogue number

This release previously used catalogue number 4326.0.30.002.