Sensitivity analysis of marginal links
Of the 1,215 individual clients who linked between the PMKeyS and SHSC data sets, 190 were identified by the AIHW as having a marginal link between the 2 data sets (a low linkage weight). The low linkage weight was primarily due to inconsistent geographical information between the 2 data sets (despite a match on the SLK), and in a few cases, slight differences in the SLK despite consistent geographical information. Sensitivity analysis was conducted to determine if marginal links should be included in the final analysis data set. Analysis revealed there were no substantial differences between the results if individuals who were marginally linked were included.
The demographic profile of ex-serving individuals who were marginally linked was also compared with the remainder of the analysis data set. The only difference observed was in relation to rank—there was a somewhat higher proportion of officers among ex-serving individuals who were marginally linked, compared with those not considered marginal links. This difference had no substantial impact on the results, so the individuals who were marginally linked were included in the final analysis data set.
Combining data across support periods within a financial year
Clients with more than 1 support period during a financial year may present with different characteristics in these different support periods. For example, their main reason for seeking assistance may be related to accommodation reasons in their first support period, and related to financial reasons in their second. Where there are multiple responses to a question, some information presented is based on the client’s response to the question when they first became a client of an agency in that financial year. Other information is based on a counting methodology that analyses the client’s responses and, if responses differ across support periods, determines the response provided most often, and the client’s longest support period for each month in the financial year. The methodology is conceptually based on the client’s journey during the course of the financial year, regardless of their movement between jurisdictions.
Combining data across financial years
For many analysis variables, data are reported for single financial years, and for the combination of financial years (usually 2011–12 to 2016–17). The rules for combining data across multiple financial years in this analysis are consistent, where possible, with other published data for the SHSC.
In this report, when comparing the age of ex-serving ADF SHS clients with ex-serving ADF members who were not SHS clients, age is calculated either as at 30 June 2017 (end of the reporting period) or as at the date of discharge.
When age is reported as at the start of support, age is as at 31 December of the financial year in which a client first received SHS support. Please note this differs to the calculation of age in National SHS reporting where the age of the client is defined as the client’s age on the start date of their first support period in the reporting period. In National SHS reporting, those who were ongoing clients at the beginning of the reporting period, the client’s age on the first day of the reporting period is used.
Using age from PMKeyS
Both the SHS and PMKeyS data sets contain information on client age. There is strong evidence of the use of default birthdates of 01/01/YYYY in the SHSC data used in the linkage for this study, as the frequency of this combination of day and month was close to 10 times the expected frequency.
Sensitivity analysis was conducted to compare the age calculated using the PMKeyS data set to age calculated using the SHSC data set. The analysis found that, while there were differences in the results produced using the 2 age measures, the differences were not substantial. As such, client age was calculated from the PMKeyS data as it was considered to be the most accurate source for the demographic information about ex-serving ADF SHS clients.
This report presents age using the following age groups: 17–24, 25–34, 35–44, 45 and over. The age groupings used in this report aim to provide consistency with Specialist homelessness services 2016–17, while ensuring the confidentiality of individuals in the data. Due to the limited number of ex-serving ADF members accessing SHS in the younger and older age groups, the groupings used in this report are broader than those used in Specialist homelessness services 2016–17. The lower bound of 17 years was chosen as the youngest individual in the linked PMKeyS–SHSC data set was aged 17 at 31 December in the financial year in which they first received SHS support. It is important to note that while the age groupings have been made as consistent as possible with those used in other published SHSC reporting, age in this report is based on information from the PMKeyS data.
When reporting across the 6-year period between 2011–12 and 2016–17, a client’s housing situation is determined as their housing situation in the first support period in the first financial year in which they received SHS support. When percentages of housing situation are presented, the denominator for the percentages is the total number of clients for whom their housing situation was known (excluding not stated).
Main reason for seeking assistance
When a client attends a SHS agency, they are asked to identify all of the reasons why they were seeking assistance. If a client identifies more than 1 reason, they are asked to choose their main reason for seeking assistance. When reporting across the 6-year period, the main reason used for reporting is the same as the main reason recorded for a client in their first financial year of receiving support (which may be their only year of receiving support).
Reason support period ended
When reporting across the 6-year period between 2011–12 and 2016–17, the total number of clients reported against each reason is the number of clients who have ever ended a support period for that reason, between 2011–12 and 2016–17. A client may have recorded the same reason for multiple support periods in a single financial year, or in different support periods over multiple financial years, however each client will only be counted once for each reason. Percentages for this variable are calculated using a denominator of all clients, excluding those whose only reason for all of their support periods ending is not stated.
Education enrolment status, labour force status and main source of income
For these variables, a client’s status is reported at both first and last presentation. When reporting across the 6-year period between 2011–12 and 2016–17, a client’s status at first presentation will be the same as their status in the first financial year for which the client received SHS support, and their status at last presentation will be the same as their status in the last financial year for which the client received SHS support.
Mental health status
Any client who was identified as having a current mental health issue, during any support period between 1 July 2011 and 31 June 2017, is reported as having a current mental health issue during the reporting period of 2011–12 to 2016–17. Clients who were identified as having a current mental health issue during multiple financial years are only counted once in the 2011–12 to 2016–17 reporting period.
This question was introduced to the Specialist Homelessness Services Collection in the 2013–14 reporting year. Clients who only received SHS support before 1 July 2013 are not included in the analysis of disability status.
The approach for assigning disability status across the 4 year period from 2013–14 to 2016–17 is as follows:
- Any client who identified as having a disability between 1 July 2013 and 30 June 2017 is assigned to Yes (having a disability).
- If a client’s disability status was Not stated for every year for which they received SHS support between 1 July 2013 and 30 June 2017, they are assigned to Not stated.
- If a client reported a combination of No disability and Not stated between 1 July 2013 and 30 June 2017, they are assigned to No (No disability).
Services needed, provided and referred
When clients present to a SHS agency, an agency worker identifies their needs. Information about a client’s needs is also updated each month while a client is still in contact with the agency. The SHSC also collects information about which services the SHS agency provided to the client. If the agency is unable to provide the needed services directly (or cannot fully meet the client’s needs) they can refer the client to other SHS agencies or other organisations that can provide those services. It is important to note that although information about clients referred for other services is collected in the SHSC, it is not possible to know if these referrals resulted in those specific services being provided to the client.
In this report, a client is recorded as needing a service if they are recorded as needing that service in any collection month or year, regardless of the number of months or years over which this need was recorded, or the number of times they presented with this need during the 2011–12 to 2016–17 reporting period. If a client is recorded as needing a service more than once during the reporting period, they are only counted once for each service. This approach is the same for services provided or referred.
Clients are recorded as having a service provided to them if they were provided with this service on at least 1 occasion during support, regardless of whether they were referred to another organisation for this service at any point during support. However, a client is only recorded as having been referred for a particular service in this report if they did not also have this service provided to them on any occasion during the support.
Clients are included in the category of Neither if the service was needed by the client and the client was neither provided with the service nor referred to another organisation for that service.
The denominator used for calculating percentages of clients who needed a service is a unique count of all clients. The calculation of percentages for a service provided uses the denominator of all clients recorded as needing that service across the 6-year reporting period. The approach is the same for calculating percentages for services referred and the category of neither referred nor reported (Neither).
Operational experience refers to 4 broad categories of deployment or operations: warlike operational experience relates to warlike/active service deployments, non-warlike operational experience relates to non-warlike deployments (for example, peace keeping, peace monitoring, United Nations assistance missions), overseas operational experience relates to humanitarian/disaster relief (International) or border protection deployments, and domestic operational experience relates to deployment of Defence aid to the civilian community. Individuals with at least 1 type of operational experience are counted as having ‘any operational experience’, and those with no operational experience are counted in the category ‘no operational experience’. Only operational experience since 1 January 1999 has been consistently identified across the 4 operational experience categories. To ensure comparability, analysis of operational experience includes only personnel hired on or after 1 January 1999. Analysis by operational experience may indicate both protective factors (such as increased training, experience and resilience) and risk factors (such as physical injury, exposure to high stress and traumatic environments and events). Clients are considered to have any operational experience across the reporting period if they are recorded as having operational experience in any of the individual financial years across the reporting period.
Days of support
Days of support is calculated by adding the total SHS support days a client received during the 2011–12 to 2016–17 reporting period, in any State or Territory in Australia. This includes both accommodation and other types of support. The total number of days of support does not necessarily represent a consecutive number of days the client received support. For example, a client who received support for 7 days may have had 2 separate periods of support: 1 for 5 days and another for 2 days.
Nights of accommodation
The length of accommodation for a client is calculated by adding each night of accommodation (short-term/emergency, medium and long-term accommodation) provided across all support periods during the 2011–12 to 2016–17 reporting period, in any state and territory in Australia. This means the total nights of accommodation does not necessarily represent a consecutive number of days the client received accommodation. For example, a client who received accommodation for 7 days may have had 2 separate periods of accommodation—1 for 5 nights and another for 2 nights.
The amount of financial assistance provided to a client is calculated by adding all financial assistance received by the client during the 2011–12 to 2016–17 reporting period. This financial assistance includes financial assistance, material aid, brokerage and vouchers provided to, or on behalf, of the client during the reporting period.