Transcript: Introducing Building a New Life in Australia

AIFS Seminar: Introducing Building a New Life in Australia: The longitudinal study of humanitarian migrants - 16 June 2015

Audio transcript (edited)

The following audio presentation is brought to you by the Australian Institute of Family Studies as part of our monthly seminar series in which we showcase national and international research related to the family. The seminars are designed to promote a forum for discussion and debate. They are open to the public and free of charge.

Dr Ben Edwards and Dr Diana Smart

Dr Edwards:
My name is Ben Edwards and I'm the Executive Manager of Longitudinal Studies at the Australian Institute of Family Studies.  I'd just like to acknowledge the Wurundjeri people of the Kulin nation, their Elders past and present, who are traditional owners of the land on which we meet. 

So in many ways this is a bit like a birthday or the birth of a child in some ways because the data will be released in the next month or so and so what we want to do is tell the story of - the pre-birth conception story I guess of this new national longitudinal study, Building a New Life in Australia, the longitudinal study of humanitarian migrants.  I'm going to be giving a sort of outline of the rationale and then a brief description of the sample and some of the study methodology characteristics and Diana is going to tell you about some of the really interesting facets and challenges that we had to face in conducting such a study.  It's been a very challenging study and I think the outcomes speak for themselves in terms of the cooperation rate and the quality of information that we've collected.  And then lastly I'll be talking a little bit about some of the technical aspects of survey weighting and also the data itself.

Firstly though is just to say that any mistakes are our own and they don't represent those views of the Australian government or the institute.  So firstly is an acknowledgement.  So the Department of Immigration and Citizenship commissioned a study in 2012 and with machinery of government changes with the election of a new government, we've moved across to the Department of Social Services.  Without their support, the study wouldn't have been possible.  The institute has been managing the study and many of the team and Diana have been managing it for the majority of the time and are in the room today.  And also we've got representatives from Colmar Brunton Social Research who've done a terrific job in conducting the survey and you'll see some of the challenges and they've been working in conjunction with multicultural marketing and management.

Like all large-scale studies, it's a team effort.  To give you a bit of a background, many of you will be aware that humanitarian migrants are fleeing trauma and persecution, many have spent many years in camps, in refugee camps or detention centres and the journeys that people take to come to Australian can be varied and difficult and we really need to know more and have more solid evidence so we can support better policies and programs.  This is a really dynamic area both politically in terms of policies but also in terms of the nature of who are the core countries from which people come to Australia, so having up-to-date, accurate evidence is really critical.

So we really need to understand the factors that aid or hinder successful settlement and we want to provide an evidence base for the research community and for public policy and service development and that's the aims of the study.  So in terms of why the study was established, there have been some longitudinal studies of immigrants in Australia in the past.  So over a decade ago there was LSIA 2 and prior to that LSIA 1, both of which had a humanitarian migrant proportion of the sample.  The proportion wasn't that large and the sample of humanitarian migrants wasn't that large and the time in which these new immigrants to Australia were followed wasn't that long so the most recent ones, they're only followed for 18 months.  So that's one of the things that distinguishes this study in that we're specifically focused on humanitarian migrants and we're following people for up to five years at this stage.

And as I mentioned before, it's a dynamic - the sources of migrants to Australia, humanitarian migrants to Australia, have changed.  So 15 years ago, people from Vietnam, Yugoslavia, Iraq.  Ten years ago, Bosnia and Croatia and Iraq and then more recently Afghanistan, Iraq and Sri Lanka.  So whilst the overall size of our humanitarian migrant intake hasn't changed that much, there's some things that have changed and one of the changes in the initiative for the study was the growth of onshore arrivals referred to - commonly referred to as illegal maritime arrivals.  That was one of the other reasons why this survey was established, to try to understand whether the pathways were different or varied for that group as well.

So just to give you a brief overview of the study survey methodology.  So it's a five-year longitudinal study, there are annual data collections but they vary.  So the initial baseline survey was conducted in the home and then a telephone interview the year later, a shorter survey, and then another interview in the home which will be happening later on this year and then a telephone survey the year after and then finally a home visit at the time in which people will be eligible for citizenship.  So we're spanning the first five years of people's lives in Australia.  We're following individuals or multiple members from families from their early months in Australia to their eligibility of citizenship and they're from across all states in Australia and in urban and regional centres.

In terms of the background for those who are not aware of the policies around this, there are two main pathways by which people can get to Australia, offshore and onshore.  So offshore, mainly they're settlements to people who are overseas through the UNHCR who are identified and referred and the visa subclasses are 200 to 204.  The technical points (indistinct) for the first forthcoming slides.  Onshore pathways are for those who want to apply for protection after they've arrived in Australia so that's commonly people who've arrived by boat or by plane and this is visa subclass 866.

In terms of the recruitment process for the study, we were lucky to have access to administrative data flows from the Department of Immigration and Citizenship, their settlement database, and this enabled us to sample from that database and to approach people.  So Colmar Brunton sent out postcards and letters prior to the start of the fieldwork and they obtained consent through the principal applicants.  So the principal applicant is the person who is the main applicant for a humanitarian visa and on that there might be multiple people applying who are referred to as secondary applicants and we refer to that group as the migrating unit.

What we did for the survey is translate survey material into 14 different languages, which was a huge challenge in terms of time but also for a variety of different reasons, and Diana is going to talk about that a little bit later.  The interviewers have to be bilingual so Colmar Brunton had to recruit and retain high quality interviewers who were able to conduct social research and ask quite sensitive questions and the training and the nature of the survey sample meant that it was quite a challenging group.  And then another novel feature of our recruitment process was the community engagement officers who are people who are respected in the local community who could help advocate, recruit and be a sort of conduit to conduct the first survey and that was really important in those initial stages.

As you can see, the obtained sample from across Australia, we had 11 sites where we conducted interviewing.  In terms of the actual sample that was recruited, there were 2,400 - almost 2,400 individuals recruited.  The majority were from a metropolitan area, a touch over 50 - over half were male.  Principal applicants consisted of the majority of our individuals, so 1,509, and most were from the offshore pathway.  So initially we had intended to interview 30 per cent onshore.  There was a change, when we were actually recruiting the sample, of government policy in mid-June 2013 which meant that any people who arrived onshore were processed on Manus Island in Nauru and that meant that we got a lower participation rate for those groups because they weren't available basically.  We maximised that as much as possible but they just weren't available.  We had a number of secondary applicants and also adolescent applicants and I'm pleased to say that in Wave 3 we will also be collecting information about their children as well, which is a very exciting prospect.

A really diverse profile, so migrants are from 35 different countries.  The majority are from the Middle-East, Iraq, Afghanistan and Iran but there's a huge - and there's a huge number of languages spoken at home, over 49 languages spoken at home, the majority being Arabic Syrian and Persian.  Clearly there's a wide array of cultural diversity and religion in this sample and a wider range of ages too.  So we had someone who was 80 years of age who was interviewed so a very broad age range as well so quite a heterogeneous sample.

In terms of the composition of the migrating unit, there was quite a varied composition, particularly for the offshore.  For the onshore arrivals, we're talking mainly about a single person so over 80 per cent were single persons and then the second most prevalent group there was the couple families with children under the age of 18.  If we turn to the offshore arrivals, a much more complex picture.  So we have single person migrating units, that's a touch over 30 per cent, but then we have one in four - a couple of families with children under the age of 18 and then some who have children as well as other family members.  So quite complex family types there and I'll give you a little bit of time to have a look through that in terms of the household composition and I think there's some very interesting - so we've got some quite large households.  I can't recall what the largest household we - in terms of migrating unit, was it eight interviews?

Dr Smart:
It was nine.

Dr Edwards:

Dr Smart:  
In the unit.

Dr Edwards:  
Yeah, nine in the unit, so quite large things.  Now I'll pass over to Diana who's going to talk about the challenges.

Dr Smart:  
Thanks everyone for coming along today and for listening in.  It's been a privilege to be involved in this study and it's also been an exciting one and a challenging one at various times.  So what I'm going to do is I'm going to talk about some of the challenges that we faced in undertaking a study like this and then I'm going to move on to talking about what we actually measured.  So what were the topics that we measured and what some of the questions were and when I'm doing that, I'm going to give you some examples of some of the answers that people could choose but of course I won't be telling you all of the answers so just remember that it's a selected group of answers that I'm going to mention.

So as Ben has already pointed out, it's a very diverse population and also highly vulnerable as we'll see later on.  So we've found that many participants had little or no education and we had expected this and prepared for it.  We found that some had never completed a questionnaire like this in their lives and so it just was really a challenge for them and some of them had difficulty understanding the more complex concepts.  For example, we asked questions about how they'd been feeling over the last four weeks and they struggled with being able to answer whether they were feeling a certain way all the time, most of the time, some of the time or never.  So this was a challenge for some of the people in our sample.  We sort of take this for granted when we're doing our research but for a population of this kind, it was a challenge.  And there were also cultural sensitivities that we needed to bear in mind.  So we needed to respect people's cultural practices when we visited their home and we needed to think about matching the language and the gender of interviewers and participants where we could.  So these were complicating factors if you like that you really need to be conscious of when you're undertaking research such as this.

Ben has already talked about the need to translate into languages and we translated into 14 different languages for Wave 1.  We selected these languages based on an analysis of the settlement database and who was going to be available in the selected sites.  So we thought that these languages would give coverage of at least three-quarters of the people that we would be approaching and we did find that more than 90 per cent of the surveys were completed in a language other than English with Arabic the most common, followed by Persian, Dari and Nepali.

We used an innovative approach to increasing accessibility of the survey.  So as well as written translations, we had audio recordings into the 14 languages and as the audio recording was playing, the group were being highlighted on the screen and this was an aim to increasing accessibility of the survey to have an audio prompt and a visual prompt for respondents.  The survey was presented on small Surface Pro computer tablets and these would be completed by the participants themselves or they could be administered by a bilingual interviewer.  And the interviewers had two computer tablets available that they took to the home so this would mean that two people could answer the survey at once which was an asset in increasing confidentiality but also limiting the time in the home.  And before participants started on their surveys, they did a short demonstration session, which got them familiar with how the Surface Pro worked and also about the different types of responses that they would be needing to use.

So translations of the surveys were obviously really vital but it really was a complex job and there were some technical issues that our Colmar Brunton colleagues had to overcome and they did so very well.  So we found that some of the languages that we might have been interested in translating weren't supported by the software, for example Karen S'gaw, but fortunately we found that in a lot of these instances the respondents had another language that was on our translated list that they could do the survey in and if they couldn't do a survey in a translated language, then we used an interpreter to assist with the interview.

And as you'd probably expect, there were variations in whether the languages went left to right and right to left and this was tricky for the programming staff because they are not able to test the languages themselves, they're English speakers, and so we needed to have the translators involved in another step to test that what had been programmed up was actually working the way it should and so this was another element of complexity that got introduced.  And while we found the audio recordings were a great asset, there were occasionally some issues around dialects and accents for some respondents.

In a study like this, there's a need for questions that may be sensitive and upsetting for participants.  To understand the challenges that people face, we need to understand what they've gone through so we really carefully considered what we could ask and tried to limit the number of sensitive questions asked but nevertheless there were a bunch of sensitive questions.  And the topics the interviewers told us have been sensitive to participants were their experiences of past trauma, having family who are not in Australia yet but who are waiting to come, the questions on mental health and post-traumatic stress and the standard income questions that virtually all studies use.  It was important for interviewers to reassure participants that their answers would be confidential and kept private and wouldn't affect their permanent visa status or their access to services or government benefits so this was a really important reassurance that needed to be made.

Now I'm going to move on to talking about what actually is in the survey and I first wanted to say that like many other longitudinal studies, we're taking it very broadly so we're trying to look at quite a number of the factors that affect people's settlement success as well as the various ways in which settlement services and settlement success can be measured.  And most of these we are carrying forward to Waves 2 and 5.  Two to five, not two and five.  So first of all, demographic information.  So principal applicants answered quite a number of questions as did the other applicant types but a lot of this information was collected only from principal applicants.  So this was information about themselves and others, on their country of birth, age, gender, how they were related to the principal applicant and whether they were living with the principal applicant or elsewhere.  We also asked whether they were sharing their residence with other people who were not on their visa application and their marital or relationship status and whether they had other children who were not living with them and it's an important factor to know that.  And finally, we collected a history of the different countries they lived in before they came to Australia.

Another main area was housing and neighbourhood.  So here we asked whether they'd received help in finding housing such as help from settlement case workers or family or friends or community groups and so on and how helpful the assistance had been.  We also asked how easy it had been for them to find housing in Australia and barriers to finding housing such as not having a rental history or references, as you can imagine, or cost or language difficulties or even understanding the process.  We asked how often they'd move homes since arriving in Australia and why they'd chosen their current home and this could've ranged from the fact that they'd been placed there by the settlement service or it was for family or community or work reasons.

We also asked details about their current home, whether it was temporary or they were on a contract or lease or another arrangement and whether they were paying rent, a loan or a mortgage or there was no money passing hands.  And we asked about their satisfaction with aspects of the home and this centred on characteristics such as the number and size of rooms, the facilities that were in the home, whether they were close to shops, public transport or schools and lastly some fairly standard questions about their neighbourhood, whether they'd found people to be friendly, they had parks or playgrounds, it was a good place to bring up children and they felt safe in the neighbourhood.

English language proficiency was a major interest and here we asked about their English language proficiency as well as their proficiency in their own language.  So we asked - as well as asking about the main language they spoke at home, we asked how well they read and write in their main language and then we'd ask about how well - before they'd come to Australia, how well they understood English, used English themselves, read English or wrote English and we would centre those questions to ask about their English language proficiency at the time of the Wave 1 interview.  We also asked whether they'd taken English language classes since they arrived and where they'd taken classes such as the AMEP, the LLNP services, and how helpful they found the English courses they'd taken.  And we know that interpreting assistance can be (indistinct) especially at this time in early months in Australia so we asked about whether they needed interpreting assistance, had they been able to get it, for what purpose, how helpful it had been and if they haven't been able to get interpreting assistance, why not.  Was it because they didn't know how to obtain it or a service wasn't available or it was too expensive.

Another area that we looked at is other education and training that they may be undertaking and while we wouldn't have expected this to be very high in Wave 1, we would expect that as the waves continue, that this will grow.  So here we asked what the highest level of education was that they'd completed before they came to Australia and if it was 
post-secondary, what their main area of study was and qualification.  We asked whether if they had qualifications, they'd been assessed and what the outcome was and if they had a qualification and hadn't had it assessed, why hadn't they had it assessed.  And turning next to whether they were currently studying, we asked whether they were taking another course of study apart from English, the type of course being undertaken and the area of study if it had a major area of study.  And finally in this area we asked what their long term educational aspirations were, what level of education they had to achieving them.

Employment and income was a very large area as you might imagine.  There are other areas that are also quite large but employment and income is probably one of the larger ones.  But in Wave 1 a lot of the employment questions were not relevant because it was early months in Australia and not so many were in employment so we expect that as the waves go by, this will become a much more intensive area of examination.  So here we asked what their prior occupations had been before they came to Australia if they had an occupation and they'd been in paid work.  We also asked if they'd been in unpaid work, so done any type of volunteering or caring for families.  And since they'd arrived in Australia, had they found it difficult to get work and what was the total amount of time that they'd actually spent in work, in paid work, and we used standard ABS type employment questions.  So whether they'd been in paid work in the last seven days, the number of jobs they had, what their main occupation currently was, what their employment conditions were and the hours worked per week and the income received in a nutshell.

We asked how satisfied they were with the jobs that they were in at the moment and if they were dissatisfied, what the reasons were.  So was it the hours worked, not enough, too much, the pay, the distance to travel or was it a poor match for their experiences and skills.  We also asked about looking for work, so what avenues had they used, had they used Centrelink or a Centrelink case worker or community links or family or friends and how helpful these had been.  So migrants on permanent protection visas are able to access a range of government benefits and so we asked about the benefits they were receiving and finally we asked what their total after tax income was from all sources.

And to further understand their financial situation, we had questions about whether they were sending money to family or friends outside Australia or giving money to family or friends inside Australia, how much they sent or gained and how regularly.  And finally in this particular module, we looked at whether they were under financial strain, asking about whether for lack of money they'd been unable to do things like pay gas or electricity bills, pay their rent or mortgage, gone without meals or were not able to (indistinct) a home and so on.

Another large area were their pre-migration experiences and here we asked about whether they or other family members had experienced traumas such as war, physical or sexual violence, imprisonment or kidnapping, persecution, a natural disaster like a flood or a drought, extreme living conditions such as lack of food or water, before they came to Australia.  There were also questions about living in refugee camps or other types of detention, which family members were with them, whether they had adequate shelter, enough food, clean water, they felt safe and so on, and what services they'd needed but hadn't been able to get.

We also asked why they'd migrated to Australia, were they placed here, was it because they had family or friends here, or was it because there were good living conditions or job opportunities and they could choose a number of these answers from the list that we provided.  Who they knew when they came here.  We know this is a really important factor that helps people settle into a country to already have family or friends here so we asked about this and we also asked whether they still had family in another country waiting to come Australia.

Moving on to the area of physical and mental health, this was also another very large area.  So we used two standard mental health scales, the Kessler Psychological Distress Scale and a short form of the Post-Traumatic Stress Disorder Scale from the Harvard Trauma Questionnaire.  We had a series of questions on physical health so we had a general rating of their health in the past four weeks, whether physical health problems had affected their daily life or work and whether they were experiencing bodily pain.  We also asked if they had a long term physical health disability, injury or condition that had lasted 12 months or more and there were quite a number of items on the stresses that they'd experienced since they arrived such as their work or family situation, fears about their family safety and so on.  There were also global questions about how difficult they were finding life in general and how well they felt they were coping.

Self-sufficiency was another important criteria of settlement success.  So here we're looking at how well they're able to carry out basic daily tasks that will affect, you know, how they live their lives.  So we asked about things like whether they felt that they were capable of looking for a job, using public transport, getting help in an emergency, using those services, finding out about their rights and accessing government services and benefits and getting help from police.  We also asked whether they'd experienced barriers in accessing government services such as not even knowing where to get help or fears that their information wouldn't be kept private or long waiting lists.

Community support and community connections was another key area for us and here asked about how much support participants had received from their national ethnic and religious communities in Australia as well as from other community groups beyond these groups.  We also asked whether they took part in activities within their national ethnic communities and other communities such as whether they attended festivals or special days or they volunteered or were members of parent or youth groups and so on.  We asked how easy they found it to make friends in Australia, understand our ways, talk to their Australian neighbours and how even these things had been for their children as well and also whether they felt they'd been made welcome in Australia and whether they felt part of the Australian community.

People come with all sorts of different qualities and personal characteristics and we know that a sense of confidence and a sense of "You can do it" can be an important support in settling in a new country.  So we asked a number of questions here about their self-confidence and self-efficacy, things like, "I'm able to do things as well as most people, I can handle whatever comes my way".  We also asked about their levels of trust in different community groups and organisations such as people in their neighbourhood, people in the wider Australian community, the police and so on.

We asked whether they'd been discriminated, stopped from doing something, being hassled or made to feel inferior because of their ethnicities, religion or skin colour.  And if they had experienced discrimination, we asked where did it happen and we had quite a long list of things that included things like on the street, being served in a shop, on public transport and so on.  We asked whether they wanted to stay in Australia and if not, why not and where they would hope to go after Australia and did they intend to apply for Australian citizenship when they became eligible.

And the last topic that we looked were their perceptions of life in Australia.  Had their experiences lived up to their expectations on things like the quality of life, the sort of support that they'd received, the job opportunities and access to education and so on.  We asked about the aspects that they had found that made settling more difficult.  We had quite a long list that they could choose from.  These were things like they didn't know anyone, their family members weren't happy, they were homesick or they weren't used to Australian ways.  And we also asked what things had helped them settle in Australia and these included things like they already had family here, they felt welcomed by the community and they felt safe and don't forget that this is the group of humanitarian migrants who've come from very (indistinct) circumstances so feeling safe was a key issue for them.

And we ended with some general questions such as what their overall experience of settling in Australia had been so far, and this ranged from very good to very hard and their thoughts on ways to improve programs and services to future humanitarian migrants.  And so it's fair to say that this is a really valuable dataset covering a whole range of areas that we are all very interested in pursuing and we look forward to actually grappling with the data that comes out of the first wave of the study.  And I'll hand it back to Ben for now.

Dr Edwards:  
Thanks Diana.  And yes, there is a lot of data that will be available.

Sorry guys in the webinar.  I was just talking about the analysis of the non-response and the extent to which the sample matches the non-response.  And in the main it does but there are some exceptions so we had lower representation of those who were living in capital cities, those who had smaller sized families, those who were in the non-IMA visa subclass, those born in Burma, the west of Africa or - west of Asia or Africa.  And so what we've done to adjust for that is we've used this technique called survey weighting.  This particular technique that we've used is weighting.  It's quite commonly used in most large-scale survey research.

We've used the module in this data to undertake it and it's been proposed by Deming and Stephan in 1940 and basically what that means is that we can represent - we represent the population proportion based on this weighting or this weight.  So we up weight those who we've missed, where we've underrepresented, and down weight those who we've overrepresented so the proportions then get matched.  And I guess the take home for people is there's three ways.  There's one for if you're diffusing information from the principal applicant, a secondary applicant weight and an overall participant survey weight.  So just keep that in mind when you're using the data.

In terms of the structure of the data, we'll be providing more support in future around access to the data but just to get a brief handle on it, the person and period dataset, we have the individual ID or person ID and migrating unit ID, the wave ID and the applicant to give you a sense of the structure of it and then we'll have new information coming in in future waves but that's how we'll be releasing Wave 1 data.  And we're hoping to have that released in the next few months.  We will let you know through the channels that you've come to know about the seminar. 

There is a process that's yet to be fully finalised but it's most likely 99 per cent sure to be similar to all the other processes that happen with any longitudinal survey the Department of Social Services has so you can apply to have access under licence to use of the full dataset and you'll be provided with that data.  And the main requirement for that access to use is that you've got - you're competent to use the information and that you've got a legitimate reason to use it and you know, you've got a research background to use it.  So it will be made available very soon and we're very excited about that prospect and for people to have access to it.

In terms of future activities, where to from here, well the data will be released very soon for Wave 1.  It'll follow very shortly after for Wave 2 release so I anticipate that we will have a Wave 2 release if not by the end of this year, by early next year, and Wave 3 will be collected at the end of the year and early next year and we'll have a release most likely in 2016/17.  We are considering asking participants in future waves about linking information, administrative information, so we can have better access to - a better understanding of their engagement with services but we'll need to explore that, we'll need to consider the ethical implications of that.  And I have also mentioned that we have and are collecting more information about the children in a child module where we're specifically going to be asking questions of children who are of ages between 11 to 17 and/or parents about their children which is a very exciting area from my own perspective being - in doing most of my research in terms of children's development, I think it's very exciting and very important when you're thinking about trying to develop and cater services to support and protect children.

So thank you very much for attending today's seminar.  We will have some time for questions but that's the end of the seminar.  I want to make one final point of acknowledgement.  I thank the support of both departments and the support of our colleagues at Colmar Brunton Social Research.  As you can see, it's a very complex and difficult study and it's been terrific to work with them and all the advisors who've had input into the content.  The AIFS team has been fantastic and it's been a terrific - in terms of degrees of difficulty with this survey, I've been involved in longitudinal studies for a while, this is a very high degree of difficulty and they've done a terrific job.  But lastly and importantly, the participants who've generously given their time to provide us with this information.  I really hope that it provides a good evidence base for debate and policy and service formation in the country, so thanks very much.

-- end –

AIFS podcasts

Leading researchers discuss significant issues affecting Australian families.

Explore our featured podcasts


AIFS news

Get the latest news about our publications, research and upcoming events.