English language testing and the exclusion of Australian citizenship

If you ask people what it means to be Australian, one answer stands out: the ability to speak English. In this 2015 ANUPoll, 92 per cent of respondents said being able to speak English was important to being Australians.

For those who argue the Turnbull government has no political nous, the proposed changes to Australian citizenship suggest this thesis is at misguided at best. Modifying citizenship, including appealing to “values” but more substantively, by formalising an English language test, will prove to be a popular decision. But what will the broader effects be on migrants?

In his press conference, the Immigration Minister said, “There’s a significant change in relation to the English language requirement which at the moment is basic. We increase that to IELTS Level 6 equivalent, so that is at a competent English language proficiency level and I think there would be wide support for that as well.”

He’s not wrong. This is not just a significant change, it is a fundamental break with established norms. To see why, you have to understand how high a barrier IELTS 6 is for many new migrants.

In 2015, ACIL Allen evaluated the Adult Migrant English Program, or AMEP. Their report is the most up to date assessment for the English literacy of recent migrants. Unfortunately, the AMEP does not use the IELTS system so a clean comparison of new migrants scores isn’t possible. Instead, we need to translate the AMEP system – called ISLPR – into IELTS equivalents:

  • In a 2006 Senate Estimates question, the Department of Immigration said an “ISLPR 2” is approximately equal to IELTS 4 or 5.
  • In the ACIL Allen review, an AMEP service provider is quoted saying an “ISLPR 2” is equal to IELTS 4.5.
  • And finally, Dr. David Ingram, the person who invented these language testing systems, references “Universities that require IELTS 6 for entry to particular courses usually require 3 in all macroskills on the ISLPR” in his submission to the Productivity Commission’s migration intake inquiry.

From this evidence, we can infer IELTS 6, the level of English proposed by the Turnbull government to be eligible for Australian citizenship is equal to ISLPR 3.

What proportion of new migrants get a score of ISLPR 3 after completing their government-allotted 500 hours of English training in the AMEP?

None. Zero per cent.

Screen Shot 2017-04-23 at 10.59.26 PM
Screen Shot 2017-04-23 at 10.59.51 PM

Based on the enrolments of AMEP from 2004 to 2012 who completed 500 hours of training, 0 per cent of new migrants reached the level required for the new citizenship test.

As per the review, “28 per cent of AMEP clients leave the programme with 0 or 0+ on all four ISLPR elements” and 7 per cent of clients after receiving 500 hours exit at ISLPR 2 (the equivalent of IELTS 4.5).

In 2004-05, about 20,000 new migrants enrolled in AMEP, a number which grew to about 30,000 in 2011-12. 60 per cent of these new migrants in AMEP classes are women and children. Of course, many new migrants do not attend AMEP classes for a variety of reasons. In 2014-15, about 80 per cent of eligible humanitarian migrants attended AMEP, 20 per cent of eligible family migrants and 8 per cent of eligible skilled migrants.

What these figures show is somewhere north of 30,000 people each year will be ineligible for Australian citizenship. While a proportion will increase their English proficiency with time (and outside the classroom), this is a slow and grueling process according to language proficiency literature.

Using conservative estimates of AMEP enrolment trends, the rate of English proficiency improvement over time, and net migration trends, anywhere between 30,000 and 40,000 new migrants each year are highly unlikely to meet the proposed English proficiency level for Australian citizenship in their first decade of settlement.

Over time, this will generate a growing population of people simply excluded from citizenship. It’s impossible to say with any certainty what this will look like over the long-term but available evidence suggests a substantial number of people will never receive Australian citizenship.

Some people might say these new migrants will simply have to learn and this is a good incentive to get it right. A tough love approach. This argument should be called out for what it is: contentment to see people permanently excluded from our society. No voting. No standing for public office. Exclusion from public service employment in many instances. The potential for expulsion from Australia by visa cancellation.

When 35 per cent of humanitarian migrants score the equivalent of an IELTS 2 after their AMEP classes finish, this argument also amounts to specifically refusing citizenship for a proportion of refugee migrants to Australia. This is despite the fact refugees love Australia more than any other group of migrants if you judge this by the number of eligible people who become citizens.

Others might say this means we need to completely tear up English language support and training so people are given the support they need. Perhaps this is true. But we also need to recognise coming to a new country is really difficult. It’s hard enough for rich, English speaking new migrants. Think about a Sudanese single mother with four children who is illiterate in her own language. A formal English language test requiring IELTS 6 is government policy telling this woman she is not welcome as an Australian citizen. And if you think this is a handpicked example on the margins of our migration program, Australia granted 1,277 “Woman at Risk” visas in 2015-16 as part of the annual humanitarian program, for “protection of refugee women who are in particularly vulnerable situations.”

And this doesn’t even get into the issue of married couples separated by an English language test, or children who can easily pass but have to watch their parents excluded.

IELTS Level 6 is by no means perfect English. You can read this practice essay and probably scoff at the simple spelling errors highlighted. The official definition is: “Generally you have an effective command of the language despite some inaccuracies, inappropriate usage and misunderstandings. You can use and understand fairly complex language, particularly in familiar situations.”

But the practical effects are enormous. It amounts to deliberate exclusion of thousands of new migrants from Australian citizenship. Back in 2015, on the same topic, I wrote, “The worst outcome is permanent exclusion from society because barriers to entry are too high. An English language test for citizenship would be such a barrier. This exclusion would occur despite an indefinite right to remain in Australia. A tiered, broken system of residency with little long-term hope.”

While almost all Australians believe speaking English is an important part of what makes someone an Australian, is this the type of society we want to live in?

4 thoughts on “English language testing and the exclusion of Australian citizenship

  1. Hey mate, I apologise for offering such a blunt and relatively esoteric criticism, but in this, and in your Lowy piece earlier in the week, I think that you have buried your lede, to the detriment of your narrative.

    The effect is that a reader, even a relatively well informed and interested reader is left struggling for several paragraphs to figure out what your central argument is. The impact is that many readers will drift off early, rather than staying with you as you lay out the evidence to back up your thesis.

    It’s especially difficult to navigate this piece, as the reader needs to process multiple unusual acronyms (with the exception on IELTS, which is well known) before you get to your punchline.

    Seperate to this article, I find the data itself both pretty astonishing, and very difficult to comprehend in those line graphs.

    Having struggled to make myself understood in both Chinese and Thai, I am amazed that 500 hours (three months full time) does not deliver better levels of speaking/listening comprehension.

    Have I read the graph accurately to think that only around 12-14% have crossed the speaking ‘proficiency’ threshold after 500 hours?

    I’m also surprised that a higher proportion of arrivals can write before they read (though I note this changes quickly). Surely that defies logic?

    Do any other English speaking countries impose similar language requirements? Non-English (such as France, Germany, Norway)?

    Don’t people just remain on PR if they don’t get citizenship?

    I don’t think that the impact of denying citizenship is social exclusion: it has practical impacts on migrants.

    New Zealanders have only recently had barriers to citizenship removed, and we don’t suffer ‘permanent social exclusion’. Our problems are much more practical – I can’t vote (though this doesn’t bother me much), and I have only limited access to the public health system (despite paying the Medicare Levy – this gets up my nose). I don’t profess to have any expertise in the NDIS, but I understand there are many significant loopholes preventing non-citizens (especially children) with disabilities from accessing support available to Australians.

    The more concerning issue for long term migrants without citizenship is the possibility of being deported for crimes and infractions (real or alleged) that would result Australians only being detained. But it’s very hard to see how 99% of Australian citizens would have a problem with this…

    Sent from my iPhone


  2. Thanks for the feedback Cam. I struggle with narrative. My excuse is I learnt how to write within the bureaucracy but I’ll try and remember to think about this stuff more in the future. I like context, recognising it’s not perfect for keeping people engaged.

    On the data. Yes, at first blush it is surprising English proficiency isn’t higher after 500 hours. However there is no information on prior literacy, socio-economic determinants, and how much practice people are getting outside the classroom. All of these will have substantial effects. Not sure about the specific quirks of writing vs. reading but they were the only good statistics in the consultant report showing the point I was trying to make. I wouldn’t use yourself as a comparison group for the average new migrant to Australia who is enrolled in the AMEP as you are an outlier in terms of education, employment and literacy demographics.

    Yes people stay on PR if they don’t get citizenship. But I think this is beside the point. What is the purpose of citizenship? What are the arguments for excluding people implicitly through English tests? You are right there are practical elements however people with permanent visas receive Medicare (after an exclusion period) and they are eligible to receive the NDIS (however I’m not across the details). Kiwis are excluded from NDIS because technically the subclass 444 visa which everyone arrives on is a temporary visa with indefinite right to remain.

    I agree it’s concerning people will be subject to deportation and I agree most people wouldn’t give a shit. But this is the reason why I think a social exclusion argument is much stronger than alternative arguments. I think the numbers have the potential to be large enough in the next decade or so for this to have electoral possibilities (obviously not the migrants who cannot vote but their broader communities).

  3. I think it is important data point to know that students in the AMEP are not entitled to English classes when they reach ISLPR 2. They cannot access SEE programs once they have this 4.5 IELTS functional English. If they don’t have PR, they have to pay international student rates for TAFE or Uni English programs. They are effectively on their own.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s