Australian Bureau of Statistics
Celebrating the International Year of Statistics 2013

FILE STRUCTURE


FILE STRUCTURE
WEIGHTS AND ESTIMATION
STANDARD ERRORS
NOT APPLICABLE CATEGORIES AND SPECIAL CODES
POPULATIONS


FILE STRUCTURE

The 2011 CEaCS Expanded CURF contains four files with confidentialised records. These files provide records for the following counting units:

1. Income unit
2. Income unit care
3. Child
4. Child care

Some of the levels for the CEaCS are in a hierarchical relationship. That is, each child at the child level of the survey is a member of an Income Unit. For the CEaCS, information was collected from income units comprising parent(s) and up to two children per family aged 0–12 years. While the word 'family' is often used interchangeably with 'income unit' in the CEaCS, the income unit may not include all members of a family e.g. an older child with an income.

Each income unit has records for the child care used by all children in the income unit for last week and usually (the income unit care level). This does include some information on children in the family not selected for the survey, as parents were asked for aggregated cost of care and use of care data for other children in the family.

Each selected child then has records both for child care use for last week and usually (child care level).

Income unit
The income unit level contains information about family and household composition, sources of income, parental educational qualifications, parental income, labour force status, work arrangements and the geographical location of the income unit. There are 5,670 records at the income unit level.

Income unit care
The income unit care level contains information about whether the income unit used care, the cost of care after the CCB and/or CCR and whether they claimed or intended to claim the CCB. There are 52,791 records at the income unit care level.

Child
The child level contains information about selected characteristics of each selected child and the main reasons for their use/non-use of certain child care services, parental participation in informal learning activities, unmet demand for early years learning centres, school readiness, early years learning centres attended prior to starting school and informal learning activities. The child level file contains 8,799 records.

Child care
The child care level contains information about child care use for the reference week and usually for each selected child and care type used e.g. weekly cost of care after the CCB and/or the CCR, frequency of use, whether CCB was claimed for care and main reason intends or does not intend to claim the CCB. The child care level file contains 111,489 records.

WEIGHTS AND ESTIMATION

As the survey was conducted on a sample of households in Australia, it is important to take account of the method of sample selection when deriving estimates. This is particularly important as a child's chance of selection in the survey varied depending on the state or territory in which they lived. Survey 'weights' are values which indicate how many population units are represented by the sample unit. See discussion in Survey Methodology.

There are two weights provided on the CEaCS CURF, as follows:

• HHWGT - household weight, for use with Income unit and Income unit care levels
• CHWGHT - child weight, for use with Child and Child care levels

The weight for the relevant level should be applied when deriving estimates from the CURF. It is essential to apply the appropriate weight for the required estimate, rather than just derive a count of records falling into each category. If a child or household weight were to be ignored, then no account would be taken of a child's or household's chance of selection in the survey or of different response rates across population groups, with the result that counts produced could be biased.

The application of weights ensures that:

• child estimates conform to an independently estimated distribution of the population by age, sex, state/territory and section of state, and
• household estimates conform to an independently estimated distribution of households by certain household characteristics (e.g. by number of adults and children), rather than to the distributions within the sample itself.

STANDARD ERRORS

Each record on the household level and person level also contains 60 replicate weights and, by using these weights, it is possible to calculate standard errors for weighted estimates produced from the microdata. This method is known as the 60 group Jack-knife variance estimator. When calculating standard errors, it is important to select the replicate weights which are most appropriate for the analysis being undertaken. The replicate weights are as follows:

• WHM0101-WHM0160 - use for Income unit and Income unit care levels
• WPM0101-WPM0160 - use for Child and Child care levels

Under the Jackknife method of replicate weighting, weights were derived as follows:

• 60 replicate groups were formed with each group formed to mirror the overall sample (where units from a collection district all belong to the same replicate group and a unit can belong to only one replicate group)
• one replicate group was dropped from the file and then the remaining records were weighted in the same manner as for the full sample
• records in that group that were dropped received a weight of zero.

This process was repeated for each replicate group (i.e. a total of 60 times). Ultimately each record had 60 replicate weights attached to it with one of these being the zero weight.

Replicate weights enable variances of estimates to be calculated relatively simply. They also enable unit record 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.

To obtain the standard error of a weighted estimate y, the same estimate is calculated using each of the 60 replicate weights. The variability between these replicate estimates (denoting y(g) for group number g) is used to measure the standard error of the original weighted estimate y using the formula:



where:

g = the replicate group number

y(g) = the weighted estimate, having applied the weights for replicate group g

y = the weighted estimate from the sample.

The 60 group Jack-knife method can be applied not just to estimates of the population total, but also where the estimate y is a function of estimates of the population total, such as a proportion, difference or ratio. For more information on the 60 group Jack-knife method of SE estimation, see Research Paper: Weighting and Standard Error Estimation for ABS Household Surveys (Methodology Advisory Committee), July 1999 (cat. no. 1352.0.55.029).

Use of the 60 group Jack-knife method for complex estimates, such as regression parameters from a statistical model, is not straightforward and may not be appropriate. The method as described does not apply to investigations where survey weights are not used, such as in unweighted statistical modelling.
NOT APPLICABLE CATEGORIES AND SPECIAL CODES

For some data items, certain classification values have been reserved as special codes and must not be added as if they were quantitative values. These special codes generally relate to data items such as income, cost of care and number of hours of care. For example, code 99999998 for the data items 'Weekly income of mother' and 'Weekly income of father', refers to income 'Could not be determined'.

Furthermore, most data items included on the CURF include a 'Not applicable' category. The 'Not applicable' category, 99999997, for the data item 'Weekly income of mother' refers to no source of income or no mother in family. Similarly for 'Weekly income of father'. The 'Not applicable' category, 99999997, for the data item 'Weekly income of parent(s)' refers to no source of income. The classification value of the 'Not applicable' category and other special codes, where relevant, are shown in the data item list in the Downloads tab.
POPULATIONS

The population relevant to each data item is identified in the data item list and should be borne in mind when extracting and analysing data from the CURF. The actual population count for each data item is equal to the total cumulative frequency minus the 'Not applicable' category.

Generally all populations, including very specific populations, can be 'filtered' using other relevant data items. For example, if the population of interest is 'Children aged 0-12 who use formal care', any data item with that population (excluding the 'Not applicable' category) can be used.

For example, the CURF data items 'All reasons used formal care' (ARATTFC (A-L)) or 'Main reason used formal care ' (MRATFOCC) are applicable only to children aged 0-12 who use formal care. Therefore, either of the following filters could be used when restricting a table to 'Children aged 0-12 who use formal care' only:

ARATTFC < 99 or MRATFOCC > 0

(Note: For those data items, the 'Not applicable' categories (i.e. those children aged 0-12 who do not use formal care) are codes 99 and 0 respectively and would be excluded from either population filter shown above.)