Key data quality information: Specialist Homelessness Services Collection, 2015–16
The AIHW plays a role in developing and maintaining national metadata standards. This work contributes to improving the quality and consistency of national health and welfare statistics. The AIHW works closely with governments and non-government organisations to achieve greater adherence to these standards in administrative data collections to promote national consistency and comparability of data and reporting.
One of the main functions of the AIHW is to work with the states and territories to improve the quality of administrative data and to compile, analyse and disseminate national data sets based on data from each jurisdiction.
Data Quality Statements are developed for each data set and made available on the AIHW Metadata Online Registry (METeOR).
The 2015–16 Specialist Homelessness Services Collection Data Quality Statement is available from METeOR.
The collection of additional information has improved the quality of data for clients on care and protection orders. Before 2015–16 clients recorded as having a care arrangement of either ‘parents’ or ‘other living arrangement’ were excluded. From 2015–16 these care types are now included. In addition, care and protection information provided in all collection months in all support periods are now included reflecting the monthly data collected from specialist homelessness agencies. These changes constitute a break in statistical time series and hence previous data on clients on care and protection orders are not comparable.
The 9% increase nationally in client numbers in 2015–16, and the similar increase in support, was the largest since the collection began in 2011–12. These increases were largely a consequence of increases in New South Wales. New South Wales homelessness services underwent a period of major transition in 2014–15 that affected continuity of reporting for some service providers. These issues did not affect New South Wales data for 2015–16. As outlined in the Data Quality Statement caution should be used when making comparisons of 2014–15 data with other years' figures for New South Wales or with data for other states and territories. Other jurisdictional-specific information can be found in the Data Quality Statement.
Further information on the data quality of 2015–16 SHSC data can be found in the Explanatory notes in the national and state and territory Supplementary tables.
Imputation strategy for the Specialist Homelessness Services Collection
Imputation was used in an effort to adjust 2015–16 reporting to correct for two types of error: agency non-response and data error in the statistical linkage key (SLK) which is used, among other things, to determine the number of clients serviced.
Agency non-response was adjusted for in two ways: by using an agency's own reported data to adjust for missing data (explicit imputation) and by weighting fully responding agencies to adjust for similar agencies with missing data (implicit imputation). Nineteen agencies underwent explicit imputation whilst implicit imputation was used for 58 agencies.
Invalid or missing SLK data were adjusted for by applying weights at the client level. These weights increased the counts of clients, taking into account the number of service period records with invalid or missing SLK data along with the observed distribution of the number of visits per client. There were about 16,500 support period records (approximately 3% of all records) with invalid or missing SLK data.
Homelessness status and other housing categories
All clients of specialist homelessness services are considered to be either homeless or at risk of homelessness. Homelessness and at-risk status is determined by the specific criteria described below. Clients who did not provide sufficient information to make this assessment are excluded.
These categories are designed to, as far as is possible, align with the ABS statistical definition of homelessness (ABS 2012a). However, there are some key areas where alignment may not occur. The ABS definition includes people living in severely crowded dwellings and as no specific question on crowding is included in the SHSC, this group cannot be separately identified.
Also, the ABS exclude certain groups of people from the homeless count where they appear to have accommodation alternatives or where there is a clear choice about the type of accommodation (for example, people who are travelling, people returning from overseas, certain owner builder or hobby farmers, and students living in halls of residence). However, if people in these circumstances become clients of specialist homelessness agencies, they are included here as either homeless or at risk of homelessness, depending on their housing situation as reported.
Clients are considered to be homeless if they are living in any of the following circumstances:
- No shelter or improvised dwelling: includes where dwelling type is no dwelling/street/park/in the open, motor vehicle, improvised building/dwelling, caravan, cabin, boat or tent; or tenure type is renting or living rent-free in a caravan park.
- Short-term temporary accommodation: dwelling type is boarding/rooming house, emergency accommodation, hotel/motel/bed and breakfast; or tenure type is renting or living rent-free in boarding/rooming house, renting or living rent-free in emergency accommodation or transitional housing.
- House, townhouse or flat (couch surfing or with no tenure): tenure type is no tenure; or conditions of occupancy are living with relatives fee free, couch surfing.
Clients are considered to be at risk if they are living in any of the following circumstances:
- Public or community housing (renter or rent free): dwelling type is house/townhouse/flat and tenure type is renter or rent-free public housing, renter or rent-free-community housing.
- Private or other housing (renter, rent-free or owner): dwelling type is house/townhouse/flat and tenure type is renter-private housing, life tenure scheme, owner―shared equity or rent/buy scheme, owner-being purchased/with mortgage, owner-fully owned, rent-free-private/other housing.
- Institutional settings: dwelling type is hospital, psychiatric hospital, disability support, rehabilitation, boarding school, adult correctional facility, youth/juvenile justice detention centre or immigration detention centre.
The period of time a client receives services from a specialist homelessness agency is referred to as a support period. A support period starts on the day the client first receives a service and ends when:
- the relationship between the client and the agency ends
- the client has reached their maximum amount of support the agency can offer
- a client has not received any services from the agency for a whole calendar month and there is no ongoing relationship.
The end of the support period is the day the client last received services from the agency.
Calculating total length of accommodation (and total length of support)
To calculate accommodation and support length, every night (for length of accommodation) or day (for length of support) the client received support or accommodation in 2015–16 is added together. This means that the total number of days/nights presented for clients does not necessarily represent a consecutive number of days/nights the client received support/accommodation. For example, a client who received accommodation for 7 nights may have had 2 separate periods of accommodation: 1 for 5 nights and another for 2 nights.
Agency remoteness area
Agencies have been classified according to their remoteness area (RA) as defined by the Australian Statistical Geography Standard (ASGS) Remoteness Structure (ABS 2012c). The latest available version of the RA indicator (from the 2011 Census) has been developed by the ABS based on the Accessibility/Remoteness Indicator Australia (ARIA) used in the 2001 Census.
Using this classification, agencies participating in the SHSC were assigned to an RA based on their recorded Local Government Area (LGA) code.
Two concordances produced by the ABS have been used to match the LGA of agencies participating in the SHSC to RAs defined by the 2011 Census. Neither concordance is 1:1—where an agency's LGA represents a proportion of an RA, the agency is assigned to the RA with the largest representation in the LGA. Where an agency's LGA code was missing, an RA was assigned using a Postal Area Index, also developed by the ABS.
Identifying and meeting service needs
Identifying clients' needs for a service
The SHSC collects information on the needs of clients during their period of support from a specialist homelessness agency. Needs may be identified by the client and/or the service provider. Although this information is collected at the beginning of a support period, updated at the end of each month a client is supported and again at the end of each support period, each individual need is only recorded once in any collection month. For these analyses, a client need for a service is recorded if the client needed that service at any time in 2015–16. For example, a client is recorded as needing short-term accommodation if they were recorded as needing short-term accommodation in any collection month of 2015–16, regardless of the number of months over which this need was recorded, or the number of times during 2015–16 they presented with this need.
Meeting clients' service needs
There are several aspects to analysing the extent to which clients' needs for assistance are met. The first is to analyse the services provided to a client directly by the specialist homelessness agency. Where agencies are unable to provide services directly to clients or unable to fully meet the need they often refer the client to other organisations (either other specialist homelessness agencies or other organisations) that can provide those services. This information is also collected in the SHSC and is considered an important form of assistance that agencies provide, although it is not possible to know if these referrals resulted in the provision of services.
All information on services that are provided, whether referred or not, are recorded in the same way as service needs. That is, a service is recorded as provided if the client was provided that type of assistance at any time in 2015–16.
In some circumstances, an agency will not be able to either provide required services directly to clients, or refer them to another organisation—this is considered to be an unmet need. Further information about unmet needs can be found in the Unmet demand section of the report.
A client is considered as Indigenous if, at any time in 2015–16, they identified as being of Aboriginal and/or Torres Strait Islander origin.
In the SHSC, information on Indigenous status is only provided with explicit client consent to report this information. Aboriginal and Torres Strait Islander was not reported for 10% of clients in 2015–16.
Clients born overseas
A client is identified as overseas-born, if at any time in 2015–16, they identified that their country of birth was a country other than Australia.
In the SHSC, information on country of birth is only provided with explicit client consent to report this information. Country of birth information was not reported for 14% of clients in 2015–16.
Young people presenting alone
Young people are defined as clients aged 15–24 who presented alone in their first support period in the reporting period.
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. For 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.
Older people are defined as clients aged 55 or older.
The age of the client is defined as their age on the start date of their first support period in the reporting period. For 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.
Clients who experienced domestic and family violence
SHSC clients were counted as experiencing domestic and family violence if any support period during the reporting period:
- 'domestic and family violence' was reported as a reason they sought assistance, or
- during any support period they required domestic or family violence assistance.
The SHSC reports on clients who are victims of domestic and family violence. Current perpetrators of domestic and family violence who may also be receiving assistance from a homelessness agency are not able to be identified within the SHSC.
Clients with a current mental health issue
A client was identified as having a current mental health issue if they provided any of the following information:
- They indicated that at the beginning of a support period they were receiving services or assistance for their mental health issues or had in the past 12 months.
- Their formal referral source to the specialist homelessness agency was a mental health service.
- They reported 'mental health issues' as a reason for seeking assistance.
- Their dwelling type either a week before presenting to an agency, or when presenting to an agency, was a psychiatric hospital or unit.
- They had been in a psychiatric hospital or unit in the last 12 months.
- At some stage during their support period, a need was identified for psychological services, psychiatric services or mental health services.
This analysis does not include clients aged under 10.
Clients on care and protection orders
A client is identified as being under a care or protection order if they are aged under 18 and have provided any of the following information in any support period (any month within the support period) during the reporting period (either the week before, at the beginning of the support period or during support):
- They reported that they were under a care and protection order (and the care arrangement was known).
- They have reported ‘Transition from foster care/child safety residential placements’ as a reason for seeking assistance, or main reason for seeking assistance.
Clients leaving care
Clients are counted as transitioning from care arrangements if, in their first support period during the reporting period, either in the week before or at presentation:
- the dwelling type was: hospital (excluding psychiatric), psychiatric hospital or unit, disability support, rehabilitation or aged care facility, or
- their reason for seeking assistance was transition from foster care/child safety residential placements or transition from other care arrangements.
Clients who were exiting custodial arrangements
Clients are counted as leaving a custodial setting if, in their first support period during the reporting period, either in the week before or at presentation:
- their dwelling type was: adult correctional facility, youth or juvenile justice detention centre or immigration detention centre or
- their reason for seeking assistance was: transition from custodial arrangements or
- their source of formal referral to the agency was: youth or juvenile justice detention centre, or adult correctional facility.
Some of these clients were still in custody at the time they began receiving support.
Children aged under 10 identified as exiting from adult correction facilities or youth/juvenile justice detention centres have been excluded because of concerns about the quality of the data, as children aged under 10 years cannot be charged with a criminal offence in any jurisdiction in Australia. Children aged under 10 transitioning from immigration detention centres have been retained in this group.
Unassisted requests for services
Unassisted requests for services provide a measure of the number of instances where a person received no immediate services from a specialist homelessness agency. It is not a measure of the number of people who did not receive services from an agency. Numbers exclude multiple requests from the same person (at any agency) on the same day, but may include requests from the same person (at any agency) on different days.
The data are presented as a daily average of requests for services because the information that is used to create the SLK was not available for 51% of the unmet requests for service in 2015–16. Without a valid SLK, it is not possible to identify whether a person requested the same service more than once from the same agency or from different agencies on different days. Similarly, people who received services at a later date, thus becoming clients, cannot be identified where a valid SLK is not available.
Data presented in the report and in the supplementary tables are mainly based on 'clients', with some data based on 'support periods' or 'client groups' (or 'presenting units'—which identify clients who present together to a specialist homelessness agency, including clients who present alone—and receive a service). Information on clients who are homeless, at risk of homelessness or part of a group of special interest, is mostly client-level data and information on agencies, unmet demand and trends data is predominantly support period data.
Data in tables that are adjusted for non-response (agency non-response and data error in the SLK) have had a weighting methodology applied which results in estimated figures that are not whole numbers. As a result, all figures in these tables are rounded to the nearest whole number and client numbers in separate columns may not add to the figure for 'all clients' due to rounding.
Population estimates used for rates calculations
All rates in this report, including historical rates, have been calculated using population estimates based on the 2011 Census. All Indigenous rates in this report are calculated using the Indigenous population estimates and projections, based on the 2011 Census.
Crude rates are calculated using the Australian Bureau of Statistics estimated resident population (ERP) at the start of the range (for example, rates for 2011–12 were calculated using the ERP at 30 June 2011). Rates for 2015–16 data were calculated using the preliminary ERP at 30 June 2015.
Population rates were adjusted (standardised) for age to enhance the comparison between populations over time that have different age structures. Specifically, direct standardisation has been used where age-specific rates are applied to a standard population (the ERP as at 30 June 2001, unless otherwise specified). This effectively removes the influence of age structure on the calculated rate and is referred to as the age-standardised rate. In this publication direct age-standardisation has been used to compare Aboriginal and Torres Strait Islander and non-Indigenous Australians (AIHW 2011).
Rate ratios are mainly used to compare Indigenous and non-Indigenous rates and provide a measure of the level of Indigenous over-representation. A rate ratio is calculated by dividing the client rate for Indigenous Australian by the client rate for non-Indigenous Australians.
Average annual rates of change
The average annual rates of change or growth rates have been calculated as geometric rates:
Average rate of change = ((Pn/Po)^(1/n) –1) x 100
Pn= value in the later time period
Po= value in the earlier time period
n = number of years between the 2 time periods.
- Australian Institute of Health and Welfare (AIHW) 2016. Youth justice in Australia 2014–15. AIHW bulletin no. 133. Cat. no. AUS 198. Canberra: AIHW.
- Australian Bureau of Statistics (ABS) 2012a. Census of population and housing: estimating homelessness, 2011. ABS cat no. 2049.0. Canberra: ABS.
- ABS 2012b. Census data quality statement: core activity need for assistance. Canberra: ABS. Viewed 17 August 2016.
- ABS 2012c. The ASGC remoteness structure. Canberra: ABS. Viewed 17 August 2016.
- ABS 2012d. Information paper: a statistical definition of homelessness, 2012. ABS cat. No. 4922.0. Canberra: ABS. Viewed 17 August 2016.
- AIHW 2011. Principles on the use of direct age-standardisation in administrative data collections: for measuring the gap between Indigenous and non-Indigenous Australians. Data linkage series. Cat. no. CSI 12. Canberra: AIHW. Viewed 17 August 2016.