4906.0.55.002 - Technical Manual: Personal Safety Survey, Expanded CURF, Australia, 2005
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 21/04/2011 Reissue
|Page tools: Print Page Print All|
USING THE CURF MICRODATA
A detailed list of the data items and categories for the CURF has been released on the ABS website in spreadsheet format. In addition to listing each data item with output categories, the data item list includes a copy of the original data item list with revisions to data items noted for comparison purposes.
CURF FILE STRUCTURE
The 2005 PSS CURF contains separate files, arranged in a hierarchy, made up of seven levels:
In using the CURF, it is important to understand how the information was collected as this determines the types of estimates that are able to be produced from the files and which levels on the CURF this information will be found.
Person level data
Respondents provided some basic demographic information about themselves and (where applicable) their current partner, such as age, sex, country of birth, employment, education and income. This information is available on the Person level as characteristics of the respondent and (where applicable) their current partner. Additional information was also collected about their general feelings of safety and lifestyle indicators. This information is also available on the person level.
Violence levels data
Respondents provided information about whether they had ever (since the age of 15) experienced:
Detailed information was then collected about their most recent incident of each of these eight types of violence. This information can be found on the Violence most recent incident level. If their most recent incident occurred more than 20 years ago, detailed information was not collected due to difficulties associated with recalling the incident.
Where a respondent had experienced more than one incident (of each of these eight types) of violence they were asked whether they had experienced each of these eight types of violence by particular perpetrator types. This was used to determine whether they had experienced each of these eight types of violence by a:
Broad information was then collected about when their most recent incident for each of these five perpetrator types occurred. This information can be found on the Violence prevalence level. In total PSS produced 40 different prevalence rates based on whether or not a person has experienced each of the eight types of violence by the five different perpetrator types.
Where a respondent had experienced any type of violence by a partner, more detailed information was collected about violence by their current and (most recently violent) previous partner. This information can be found on the Violence partner level.
The following diagram illustrates the different types of violence contained on the Violence levels.
Abuse (before the age of 15) level data
Respondents provided information about whether they had ever experienced abuse by an adult before the age of 15. Information was collected about whether they experienced:
Where a respondent had experienced abuse by an adult before the age of 15, some broad information was collected about how old they were at the time of the first incident and their relationship to the perpetrator/s of the first incident. This information can be found on the Abuse level.
Harassment level data
Respondents provided information about whether:
Where a respondent had experienced any of these types of harassment they were then asked if it had occurred in the 12 months prior to interview. This information can be found on the Harassment level.
Stalking level data
Respondents provided information about whether they had ever experienced:
Where a respondent had experienced stalking by a man/woman, they were asked about the stalking behaviours, frequency and length of stalking, their relationship to the perpetrator/s, police involvement and fear or anxiety experienced due to being stalked. This information can be found on the Stalking level.
USING THE FILE LEVELS
This section contains:
The Person level contains one record for every respondent. The remaining six levels are in a hierarchical relationship with the Person level, and they exist to describe 'one to many' or 'many to many' relationships. A respondent may have one record, or several records, on any of these levels. Respondents who did not experience, for example stalking, will have one record on the Stalking level, which falls into the 'Not applicable' category in the index item (see below for information about index items). Respondents who experienced stalking by a male only, or by a female only, will have one record on the Stalking level, for each data item. Respondents who experienced stalking by a male and stalking by a female will have two records on the Stalking level, for each data item.
If a person had experienced:
They would have:
Information on the Person level can be used in conjunction with information on the other levels. For example, when looking at a person's experiences of stalking, data on the Stalking level, such as the type of stalking behaviours, can be cross tabulated by information on the Person level, such as the age and sex of the respondent and their labour force status. This is done by merging the person level data onto the Stalking level file. An example of the code to merge data from the Person level to a lower level is provided in Record identifiers and copying data to a lower level.
However, information on each of the lower levels cannot be merged across the levels.
The Violence levels (prevalence, most recent incident and partner) all contain information about experiences of violence. There are two types of Violence data available:
The Violence prevalence level is used to determine whether a person experienced a particular type of violence, while the Violence most recent incident level is used to assess the characteristics of the most recent incident of that type of violence. The Violence partner level is used to assess a person's overall experience of current partner and/or previous partner violence.
The three Violence levels are mutually exclusive, that is, data cannot be merged or summed across these levels.
All of the six lower levels on the CURF contain index items, which must be used when extracting data from these levels to produce estimates of the number of people with a particular experience or characteristic. The index items restrict the population to those people who experienced a particular type of violence/abuse/harassment/stalking. Where multiple index items exist on a level, all index items must be used. A working example to show why the population needs to be restricted to those who experienced a particular type of violence is provided below.
If you were trying to produce estimates about reporting to the police, what if a person had experienced sexual assault by a male and reported it to the police; and they had also experienced physical assault by a female but this had not been reported to the police? It is not possible to produce an estimate that accurately reflects this person's actions, as they both did and did not report to the police. In presenting information about the characteristics of violence, it must to be presented separately by the eight types of violence such as 'Characteristics of sexual assault by a male' and 'Characteristics of physical assault by a female'. This principle applies to all of the six lower levels.
Information available on the CURF
The following information provides an overview of the data available and the number of records on each level of the CURF.
The Person level contains information about each survey respondent including geographic, demographic and socioeconomic characteristics, as well as information on general safety, emotional abuse and whether experienced abuse before the age of 15. This level contains one record per person.
Information available on the CURF includes:
The person level also contains information about a respondent's partner, where applicable. Partner information (provided by the respondent) available on the CURF includes:
There are 16, 413 records on the Person level.
Violence prevalence level
The Violence prevalence level is for the lifetime prevalence of violent experiences since the age of 15 and is used to extract information about whether a person has ever experienced violence by the different perpetrator types. Prevalence rates are calculated by cross tabulating the type of violence experienced, sex of perpetrator and type of perpetrator. Cross tabulating these three data items results in 40 combinations or prevalence rates, for example, 'physical assault by a male stranger' or 'physical assault by a male boyfriend or date' or 'physical assault by a male current partner', and so on. This level contains between one and 40 records per person, depending on whether they have experienced violence since the age of 15 and how many types of violence they have experienced.
Information available from the Violence prevalence level includes:
There are 22,417 records on the Violence prevalence level.
Violence most recent incident level
The Violence most recent incident level contains information about the characteristics of a person's most recent experience of violence (excluding incidents that occurred more than 20 years ago), for each type of violence experienced, by sex of perpetrator. There are a possible eight combinations of violence on this level, for example, 'physical assault by a male' or physical assault by a female' or physical threat/attempted physical assault by a male', and so on. This level contains between one and eight records per person, depending on whether they have experienced violence since the age of 15 (excluding incidents that occurred more than 20 years ago) and how many types of violence they have experienced.
Information available on the Violence most recent incident level includes:
While many of the data items on this level are available for all types of violence, it is not possible to sum data for the different types of violence to produce estimates for total violence. See Index items.
There are 19,405 records on the Violence most recent incident level.
Violence partner level
The Violence partner level contains information about the characteristics of a person's experience of violence by a current partner and by a previous partner. Where a person has experienced violence by more than one previous partner, information relates to the most recently violent previous partner. This level contains one or two records per person, depending on whether they have experienced violence by a current and/or previous partner.
The following information is available for both current partner and previous partner violence:
While these items are available for both current partner and previous partner violence, it is not possible to sum current partner and previous partner data to produce a combined estimate for 'partner violence'. See Index items.
The following information is available for current partner violence only:
The following information is available for previous partner violence only:
There are 16,448 records on the Violence partner level.
The abuse level contains basic information about experiences of abuse before the age of 15, by the type of abuse (physical or sexual). This level contains one or two records per person, depending on whether they experienced physical and/or sexual abuse before the age of 15.
Information on the Abuse level includes:
There are 16,995 records on the Abuse level.
The Harassment level contains basic information about experiences of sexual harassment by the type of harassment experienced (for example obscene phone calls or indecent exposure) and the sex of the perpetrator. This level contains between one and eight records per person, depending on whether they have experienced harassment and how many types of harassment they have experienced.
Information on the Harassment level includes:
There are 25,469 records on the Harassment level.
The Stalking level contains information about experiences of being stalked by the sex of the perpetrator. This level contains one or two records per person, depending on whether they have experienced stalking by a male and/or female perpetrator.
Information on the Stalking level includes:
There are 16,540 records on the Stalking level.
Record identifiers and copying data to a lower level
Identifiers can be used on records at each level of the file to merge information from the person level to the lower levels. Each person has a unique random identifier - ABSPID, which appears on all levels of the file.
There may be instances when a data item is not contained on the level of the file required by users. For example, information about the characteristics of a person's most recent experience of violence is contained on the Violence most recent incident level, while that person's geographic and demographic information is contained on the Person level. To apply Person level characteristics to the Violence most recent incident level, the Person level characteristics (on the Person level file PSS05PER) must be merged on to the Violence most recent incident level file (PSS05VIO) using the person identifier (ABSPID). To do this, the following SAS code (or equivalent) can be used:
PROC SORT DATA = PSS05PER;
PROC SORT DATA = PSS05VIO;
MERGE PSS05VIO (IN = A)
PSS05PER (KEEP = ABSPID AGETC SEX STATEUR PSSFINW2 IN = B);
IF A AND B THEN OUTPUT; *Only keeps records which are present on both files*
The KEEP statement includes all person data items specified to be merged onto the Violence most recent incident level file. Person characteristics can be merged on to the Violence most recent incident level to identify the characteristics of the people who reported one or more violent experiences since the age of 15, and these characteristics are applied to each 'type of violence' record on the Violence most recent incident level. In the above example, age (data item AGETC), sex, State or Territory of usual residence (STATEUR) and the person weight (PSSFINW2) are merged onto the Violence most recent incident level.
Use of weights
Every record contains a 'weight'. The weight indicates how many population units are represented by the sample unit. The CURF contains 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 CURF. If the chance of selection is not accounted for, by use of appropriate weights, the results will be biased.
Each person record contains a weight. The identifier is PSSFINW2, and 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 weight, and not just counting the number of records in each category. If person 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. More information about these differences is provided in the Introduction.
Replicate weights have been included on the CURF 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 CURF contains 60 person replicate weights (WPM0201 to WPM0260).
For more information on sampling error see Reliability of estimates.
Continuous data items
When analysing continuous items on the Person level, 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, 'Gross cash income of person (weekly)' (CINCWR) has reserved values of:
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.
Multiple response items
There are a number of data items on the CURF that contain multiple responses. This means that the respondent 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:
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 'Whether services used after the most recent incident' (DVGETHP), on the Violence most recent incident level, which has six response categories. From these categories six separate data items have been produced - DVGETHPA, DVGETHPB.....DVGETHPF.
The special codes for 'Not applicable', 'Not known' and 'Refusal' are provided in the data item list (eg values of 97, 98 and 99). In some cases special codes such as the 'Not applicable' population will fall into the first SAS output category, rather than in its own category as in the example above. Where the special codes fall into the first SAS output category, that category will have:
To enable CURF users greater flexibility in their analyses, two Socio-Economic Index for Areas (SEIFAs) have been included on the CURF, as well as two 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.
For more details on SEIFA see the Information Paper: An introduction to Socio-Economic Indexes for Areas (SEIFA), 2006 (cat. no. 2039.0).
These documents will be presented in a new window.