Australian Bureau of Statistics
3416.0 - Perspectives on Migrants, June 2010
Previous ISSUE Released at 11:30 AM (CANBERRA TIME) 15/06/2010
|Page tools: Print Page Print All RSS Search this Product|
SETTLEMENT OUTCOMES FOR HUMANITARIAN PROGRAM MIGRANTS - EXPERIMENTAL ESTIMATES FROM THE MIGRANTS STATISTICAL STUDY
WHAT IS THE CENSUS DATA ENHANCEMENT PROJECT?
The Migrants Statistical Study is a component of the Census Data Enhancement (CDE) project which brings together a 5% sample of Census records with data from DIAC's Settlement Database to form a new dataset. The linking methodology used to link the SDB and the 2006 Census data was probabilistic linking. The method linked records from two files using several variables common to both files. The data was linked without the use of name and address. Records were also weighted such that estimates produced from the file would match to key SDB totals.
The establishment of the CDE project and associated details are presented in ABS information papers Census Data Enhancement Project: An Update (ABS, 2006a), Enhancing the Population Census: Developing a Longitudinal View (ABS, 2006b) and the Australian Statistician’s Statement of Intention (ABS, 2005).
WHAT IS THE DATA QUALITY?
Since only a 5% sample of the Census records are included, there will be sampling error associated with estimates from the linked file. Sampling error is the error in the estimate caused by selecting a sample rather than taking a complete enumeration of the population. It is caused by the variability in responses for different members of the population. Generally speaking, as the sample size increases, the amount of sampling error decreases. Using a 5% sample, there will be some subpopulations for which there will be too few observations in the sample to obtain reliable estimates (i.e. the sampling error is too large).
The most common way of quantifying sampling error is to calculate the standard error for the published statistic. As in the case of other ABS outputs, in this article and the associated tables, estimates with a higher level of relative standard error (RSE) have been flagged. It should be noted that relative standard errors (RSEs) only provide a measure of sampling error, they do not take into account other potential sources of error such as error introduced by the linking process. Error associated with the linking process may occur due to false or missed links. False links are where records for different individuals were linked, which may occur when different people have similar characteristics. Missed links may occur when an equivalent record cannot be found. This may happen in cases where a different value is held on each file in respect of one or more linking variables or where variables had missing values. Another potential source of error occurs where there is not an equivalent record on the other dataset, e.g. the person has a record on the SDB, but was missed in the Census.
The major quality issue is the number of SDB records in the population of interest that cannot be linked (approximately 34%). To draw conclusions from the linked data about the entire population of interest assumes that the unlinked records would have similar characteristics. In terms of the characteristics (variables) listed on the SDB (e.g. age, sex, visa category), the linked and unlinked files generally had similar proportional distributions, but there were some subpopulations that had a small degree of under-representation on the linked file. For more details on the linkage, refer to section 5 of the Quality Study paper 'Research paper: Assessing the Quality of Linking Migrant Settlement Records to Census Data' (cat. no. 1351.0.55.027).
The other issue to consider is whether or not the analysis variables from the Census (e.g. employment status) have similar proportional distributions for the linked and unlinked records, although this is hard to assess since the unlinked SDB records do not have the Census variables. It is expected that there would be some correlation between the observed SDB variables and the unobserved Census variables for the unlinked records. Thus the fact that unlinked records have generally similar distributions of SDB characteristics to the linked records, indicates that the Census variables may also be similar. The analysis in this paper is based on the assumption that relationships between variables observed in the linked data, are the same for those in the unlinked data.
While all estimates have been benchmarked to totals from the Settlements Database it should be noted that these totals include some people who have died, or people who have emigrated from Australia between the time of their arrival and Census night. Therefore, although the estimates are weighted to 'known' benchmarks, the benchmarks may not be realistic, and thus caution needs to be exercised in reporting what the estimates represent.
The linkage process uses specific variables available on both files in order to create links between records on the two files. The quality of the linkage is dependent on the quality of those variables. For some groups there are particular quality issues for certain variables. During the creation of the linked file it was found that some groups were more difficult to link than others, resulting in some degree of under or over-representation on the linked files. In general, younger, more mobile groups within the community tend to be more difficult to enumerate, and hence more difficult to link. The characteristics of the subpopulation in the sample may not be a good representation of the subpopulation as a whole. Weighting records to known benchmarks reduces this problem to some extent.
Due to the impact of sampling error and, in particular, potential additional error introduced by the linking process itself, additional caution should be used when using these results. Despite this caveat, the file does provide a good indication of key characteristics within different migrant subgroups, including a range of data not currently available from other sources.
Further information about the data and the linking methodology used in the Statistical Study is available in the Quality Study paper 'Research paper: Assessing the Quality of Linking Migrant Settlement Records to Census Data' (cat. no. 1351.0.55.027).This research paper provides a summary of the Migrants Quality Study.
WHY DID WE LOOK AT HUMANITARIAN PROGRAM MIGRANTS?
In 2008, Australia ranked as one of the top three countries contributing to the resettlement of refugees (USCRI, 2009).
Humanitarian Program migrants require specific policy responses and programs that reflect the unique circumstances of their arrival to Australia. Humanitarian Program migrants have often experienced trauma in their lives prior to their arrival in Australia, such as living in refugee camps for long periods of time. As a result, it may take longer for Humanitarian Program migrants to achieve employment and other positive settlement outcomes and they may require more assistance when they arrive because of differences between Australian lifestyle and culture and their previous experiences. However, this group may also be more motivated to find employment or further their education because they have been given the opportunity to start a new life.
Due to small population size this is also a migrant group for which little survey data are available. The most recent reliable data on Humanitarian Program entrants was collected as part of the second Longitudinal Survey of Immigrants to Australia (LSIA), from 1999 to 2002 conducted by DIAC, which is now almost a decade old. Consequently very little is known about the characteristics and settlement outcomes of this group. Issues relate to pre and post settlement experiences and situations which are often very different to those of mainstream migrants. Hence there is a need for specific data on this group.
There are two main categories in the Humanitarian visa category; Refugees and Special Humanitarian Program migrants. The onshore (asylum or protection) component offers protection to people in Australia who meet the refugee definition in the United Nations Refugees Convention. The offshore (resettlement) component offers resettlement for people outside Australia who are in need of humanitarian assistance. The 'Refugee' category is for people subject to persecution in their home country. The 'Special Humanitarian Program (SHP)' category is for people who, while not being refugees, are subject to substantial discrimination amounting to a gross violation of their human rights in their home country. For further information about these visa categories, refer to DIAC's website: http://www.immi.gov.au/visas/humanitarian/.
WHAT DO WE ALREADY KNOW?
Selected demographic characteristics from existing data sources
Some information on Humanitarian Program migrants is currently available from DIAC's Settlement Database. This data has good information on the size of the humanitarian population and some characteristics of this population, so it has been used in this article to give an indication of age group, sex and country of birth.
Based on analysis of the SDB file, of the 806,952 migrants in scope for this study, 76,719 (10%) were Humanitarian Program migrants. There were more male Humanitarian Program migrants (55%) than females (45%). In contrast, there were more females (52%) than males (48%) in the total population of migrants, which was largely due to a high proportion of female migrants in the family visa category (61%). In the 2006 Census, there were more females (51%) than males (49%) in the population of all persons (ABS, 2007).
Migrants, Age by Sex - 2006
Note: Total migrants includes Humanitarian Program migrants.
Source: Visa subclass, Age and Sex (DIAC, SDB extract)
Humanitarian Program migrants tend to be younger when compared to the total migrant population. In 2006, 39% (30,233) of Humanitarian Program migrants were less than 20 years of age, which was almost double the proportion in the total migrant population (22%).
Half of Humanitarian Program migrants were born in the region 'North Africa and the Middle East' (50%), which is higher than the proportion of the total migrant population born in this region (9%). Migrants born in Iraq, Sudan and Iran, comprised almost half of the total humanitarian intake (46%). A majority of the Sudanese migrants (71%) and almost half of Iraqi migrants (48%) were granted visas under the Special Humanitarian Program. In contrast, most migrants born in Iran were Refugees (57%).
By way of comparison, the most common regions of birth for skilled migrants were 'North-West Europe' (23%) or 'North-East Asia' (18%) while family migrants were commonly born in 'South-East Asia' (21%) or 'North-West Europe' (20%).
Migrants are either primary or secondary applicants. The primary applicant must satisfy the primary criteria for the grant of a visa under the Migration Regulations. A secondary applicant is a member of the family unit of the primary applicant, i.e. their spouse, an interdependent partner, a dependent child or a dependent relative of the primary applicant. Almost two-thirds (63%) of Humanitarian Program migrants are secondary applicants compared to 53% of skilled migrants and 13% of family migrants.
WHAT DID WE FIND?
Selected demographic characteristics from the linked file
The linked file has increased our capacity to investigate characteristics and settlement outcomes of Humanitarian Program migrants as a group. The remainder of this article will focus on data from the linked file, which will provide information on various demographic characteristics, including: State of usual residence, religion, family composition and English proficiency. Information on various settlement outcomes is also presented, including: education, employment, occupation, income and volunteer work.
State of Usual Residence
Most Humanitarian Program migrants lived in either New South Wales (40%) or Victoria (30%) at the time of the 2006 Census. A small proportion of Humanitarian Program migrants lived in Tasmania (2%), the Northern Territory (1%) and the Australian Capital Territory (1%).
The religious affiliation of Humanitarian Program migrants as reported in the 2006 Census was predominantly Christian (53%) and Islam (33%). In comparison, 48% of the total migrant population recorded their religious affiliation as Christianity, but only 10% recorded Islam. Only a small proportion of Humanitarian Program migrants indicated that they didn't have a religion (2%) compared with 19% of the total migrant population.
Humanitarian Program migrants may live in different family groups when compared to other migrants due to their wide range of cultures and backgrounds.
Half (50%) of Humanitarian Program migrants over 15 years of age were married and 41% had never been married when the 2006 Census was conducted. Almost two-thirds (64%) of the total population of migrants, aged 15 or older, was married. A slightly higher proportion of Humanitarian Program migrants were widowed (4%) than the proportion in the total migrant population (2%).
Most (61%) Humanitarian Program migrants lived as a couple family with children in 2006. There was a higher rate of single parents in the Humanitarian Program migrant population (18%) in comparison to the total migrant population (7%). On the other hand, there were less Humanitarian Program migrants living as a couple with no children (6%) than there were in the total migrant population (22%).
Humanitarian Program women over the age of 15 were more likely to have larger families than other migrant groups. For example, there was a higher rate of females who had 3 or more children in the Humanitarian Program migrant population (31%) when compared to the total migrant population (12%).
For each person who speaks a language other than English at home, the 'Proficiency in Spoken English' variable on the 2006 Census classifies their self-assessed proficiency in spoken English. Responses to the question on proficiency in English were subjective. For example, one respondent may have considered that a response of 'Well' was appropriate if they could communicate well enough to do the shopping while another respondent may have considered such a response appropriate only for people who can hold a social conversation. Proficiency in spoken English should be regarded as a broad indicator of a person's ability to speak English rather than a definitive measure of his/her ability.
English proficiency was likely to be lower for Humanitarian Program migrants than for other migrant groups because most of them were not born in English speaking countries. The data showed that only 5% of Humanitarian Program migrants spoke only English at home, compared to a third (33%) of the total migrant population.
Over a third (36%) of Humanitarian Program migrants reported speaking English well. However, a quarter (25%) of Humanitarian Program migrants said they didn't speak English well, which is higher than the proportion of total migrants in this category (11%). Comparably, 8% of Humanitarian Program migrants did not speak English at all, which was higher than the proportion of total migrants in this category (3%).
Year of arrival refers to the year of arrival recorded on DIAC's settlement database for clients with a permanent residency visa. For an overseas application, the arrival date is the first arrival date after the visa is granted. For an onshore application, the arrival date is the arrival date immediately prior to the grant (this arrival might have been on a temporary visa).
Most recent Humanitarian Program migrants were Australian citizens three years after their arrival date. Relatively few of those who arrived two years prior to the 2006 Census in 2004 had received Australian citizenship status (16%), however, when migrants had been in Australia for three years (2003), the proportion with Australian citizenship increased to 69%. Within the same timeframe, the rate of Australian citizenship for skilled and 'family and other' migrant groups did not increase as markedly, so that less than a third of skilled migrants (29%) who had been in Australia for three years (since 2003) had received Australian citizenship status and only 16% of 'family and other' migrants. Prior to 2007, the residency requirements for becoming an Australian Citizen included a 2 year period living in Australia as a permanent resident in the 5 years immediately before applying, including 1 year in the 2 years immediately before applying. For further information about citizenship requirements, refer to DIAC's website: http://www.immi.gov.au.
Almost half of all Humanitarian Program migrants were full-time students (46%) at the time of the 2006 Census. This was higher than the proportion of all migrants studying full-time, which was just over a quarter (26%). Skilled migrants (30%) and family migrants (16%) were less likely to be studying full-time when compared to Humanitarian Program migrants.
Most (62%) secondary applicant Humanitarian Program migrants were full-time students, which was greater than the proportion of primary applicant Humanitarian Program migrants who were full-time students (20%).
In 2006, 84% of the migrant population was over the age of 15. The data in this section includes prior education level, current employment and participation in voluntary work and it is restricted to migrants aged 15 years and over.
The proportion of Humanitarian Program migrants who had completed year 12 or equivalent (47%) was lower than the proportion in the general migrant population (75%). There was a higher proportion of Humanitarian Program migrants (13%) with an educational level of year 8 or below when compared to the general population of all migrants (3%). The rate of persons who never attended school was higher for Humanitarian Program migrants (7%) than it was for the total migrant group (2%).
In 2006, a higher proportion of skilled and family migrants had completed a postgraduate degree (18% and 7% respectively), when compared to their humanitarian counterparts (1%). This was also the case for those migrants who had completed bachelor degrees. A third of skilled migrants had completed a bachelor degree (34%), compared with 22% of family migrants and 5% of Humanitarian Program migrants.
A higher proportion of Humanitarian Program migrants recorded 'not applicable' for their level of non-school qualification (61%) than the total migrant population (32%). The 'not applicable' category included persons who have a qualification that is out of scope of the classification, persons with no qualifications and persons still studying for a first qualification.
Labour force status is the primary measure for assessing employment outcomes of Humanitarian Program migrants.
In 2006, a larger proportion of family and skilled migrants over the age of 15 were employed full-time or part-time in comparison to Humanitarian Program migrants. Over a third of migrants in the family stream were employed full-time (36%) and half of skilled migrants were employed full-time (50%), whereas only 17% of Humanitarian Program migrants were working full-time. In the 2006 Census, over a third (37%) of the total Australian population was employed full-time (ABS, 2007).
Humanitarian Program migrants were more likely to be in the group 'not in the labour force': 57% were in this category compared to 36% of migrants in the family stream and 23% of skilled migrants. In comparison, Census figures from 2006 indicate that a third (33%) of the total Australian population aged over 15 are 'not in the labour force' (ABS, 2007).
Both male (63%) and female (35%) skilled migrants, aged over 15, were more likely to be employed full-time when compared to the proportion of Humanitarian Program migrants employed full-time (25% of males and 6% of females).
Skilled main applicants (63%) and skilled secondary applicants (32%) were more likely to be employed full-time when compared to the proportion of Humanitarian Program migrants employed full-time (23% of main applicants and 10% of secondary applicants).
In general, the proportion of Humanitarian Program migrants in employment increased as the number of years since their arrival in Australia increased. For example, 37% of Humanitarian Program migrants who had arrived five years before the 2006 Census were employed, compared to 21% of recent arrivals who had only been in Australia since 2005.
A higher proportion of Humanitarian Program migrants who settled in Western Australia were employed (38%) than those who settled in Victoria (28%).
The most common level of post school qualification for employed Humanitarian Program migrants, aged 15 years and over, was a 'Certificate' level course (19%), followed by 'Advanced diploma or diploma' courses (12%) and 'Bachelor degree' (9%). However, almost half of employed Humanitarian Program migrants (48%) didn't have a post school qualification.
In the 'Occupation' and 'Income' sections, the comparisons were made between skilled migrants and Humanitarian Program migrants. This comparison is interesting as it allows us to compare those who have migrated to Australia on the basis of their specific skills with Humanitarian Program migrants who have not been selected on this basis and do not necessarily have the English proficiency level or educational qualifications required to meet skilled selection criteria.
Almost a third (31%) of Humanitarian Program migrants were employed as labourers and 18% were employed as technicians and trades workers. A higher proportion of skilled workers were employed in professional occupations such as managers (11%) and professionals (36%) when compared with Humanitarian Program migrants (4% and 6% respectively).
In general, Humanitarian Program migrants, aged 15 years and over, earnt less per week than skilled migrants. There was a higher proportion of skilled migrants (49%) earning pay ranges above $600 per week in comparison to Humanitarian Program migrants (12%).
One barrier migrants may face when trying to secure employment in Australia is their level of English proficiency. In 2006, almost three quarters (74%) of Humanitarian Program migrants who did not speak English well or did not speak English at all were 'not in the labour force' and only 16% of this group were employed. In contrast, 47% of Humanitarian Program migrants who spoke only English at home or who spoke English very well were 'not in the labour force' and 40% were employed.
Volunteer work is one option for Humanitarian Program entrants to gain experience in the local job market, which may increase their chances of gaining paid employment. In the 12 months prior to the 2006 Census, 9% of Humanitarian Program migrants, aged 15 years and over, participated in volunteer work through an organisation or group. In comparison, 13% of the total migrant population participated in voluntary work over this period.
In terms of other unpaid work, the 2006 Census also provided details of unpaid assistance to a person with a disability, the provision of unpaid child care and the number of hours spent on unpaid domestic work. These items are an additional way to measure economic contribution to society because it may reduce burden on community services, such as child care centres, hospitals, aged care facilities, etc.
A higher proportion of Humanitarian Program migrants provided unpaid assistance to a person with a disability (11%) in the two weeks prior to the 2006 Census when compared with the total migrant population (7%). In contrast, a lower proportion of Humanitarian Program migrants provided child care in the two weeks prior to the 2006 Census (27%) when compared to the total migrant population (36%).
A higher proportion of Humanitarian Program migrants (41%) didn't spend any time on unpaid domestic work for their household in the week prior to the 2006 Census in comparison to the proportion of total migrants who didn't spend any time on these chores (24%).
WHAT IS THE IMPACT OF LENGTH OF RESIDENCY?
There were some differences between recent Humanitarian Program migrants and Humanitarian Program migrants who had lived in Australia for a length of time between 4 and 6 years. Recent migrants were more likely to be younger with a median age of 20 and not speak English well or at all (38%) when compared to longer term residents who had an a median age of 29 and a quarter (25%) didn't speak English well or at all. On the other hand, it was more common for those aged 15 years and over who had lived in Australia longer to be employed (38%) and have a post school qualification (30%) when compared to recent migrants (22% and 22% respectively). FURTHER INFORMATION
If you have any queries or comments about this article, or you would like to obtain further information about migrant and ethnicity related statistics, please visit the Migrant and Ethnicity Theme Page on the ABS website or contact the National Migrants Statistics Unit directly by email on firstname.lastname@example.org.
LIST OF REFERENCES
ABS 2005, Australian Statistician’s Statement of Intention.
ABS 2006a, Census Data Enhancement Project: An Update, June 2006, cat. no. 2062.0, ABS, Canberra.
ABS 2006b, Discussion Paper: Enhancing the Population Census: Developing a Longitudinal View, 2006, cat. no. 2060.0, ABS, Canberra.
ABS 2007, 2006 Census QuickStats: Australia, ABS, Canberra.
DIAC Citizenship, viewed 15 April 2010, <http://www.citizenship.gov.au/learn/law-and-policy/legis_changes/res_req_changes/>.
DIAC Settlement Reporting Facility, viewed 15 April 2010, <http://www.immi.gov.au/living-in-australia/delivering-assistance/settlement-reporting-facility/>.
DIAC website, viewed 15 April 2010, <http://www.immi.gov.au/visas/humanitarian/>.
USCRI, World refugee survey 2009, Tables and graphs, Resettlement by country, viewed 9 March 2010, <http://www.refugees.org/FTP/WRS09PDFS/Resettlementbycountry.pdf>
These documents will be presented in a new window.
This page last updated 20 December 2011