FactCheck: do 'over a million' people in Australia not speak English 'well or at all'?

A growing number of people in Australia cannot speak English well or at all, over a million people.

– Senator Pauline Hanson, Senate speech, September 19, 2018

One Nation Party leader and Senator for Queensland Pauline Hanson is urging a rethink on Australia’s immigration policy, including changes to the “number and mix” of migrants coming to the country.

In a Senate speech, Hanson outlined a number of concerns she has with what she described as Australia’s “failed immigration policy”, including issues with social integration and the establishment of “culturally separate communities”.

The senator said a “growing number of people in Australia cannot speak English well or at all, over a million people”.

Is that right?

Checking the source

In response to The Conversation’s request for sources and comment, an advisor to Senator Hanson accurately cited Census data showing the number of people who self-reported they spoke English “not well” or “not at all” was 820,000 in 2016, up from 655,000 in 2011 and 560,000 in 2006.

To reach a calculation of “over a million people” in 2018, Hanson’s office:

• added 66,000 people to the 2016 Census results, based on the assumption that the growth in the number of people in this category would be the same between the 2016 and 2021 Census as it was between 2011 and 2016, and

• added a further 149,294 people to the 2016 results, based on the assumption that 10% of the 1,492,947 people who didn’t respond to the question in the Census about language proficiency did not speak English “well or at all”.

You can read the full response from Hanson’s office here.

Verdict

Senator Pauline Hanson said “a growing number of people in Australia cannot speak English well or at all, over a million people”.

The most up to date information available on this question comes from the 2016 Census. The data show that the number of people who self-reported speaking English “not well” or “not at all” in that year was 820,000.

Hanson was correct to say that number has been growing, from 560,000 people in 2006 to 820,000 people in 2016. This amounts to a rise from 2.8% of Australian residents in 2006 to 3.5% in 2016.

Over the same time, among people who speak a language other than English at home, the percentage of people who self-reported speaking English “not well” or “not at all” fell, from 17.5% in 2006 to 16.6% in 2016.

It’s important to keep in mind that self-reporting is not the most accurate measure. Some people will over-estimate their language capabilities, while others will under-estimate theirs.

What do the data show?

In its five-yearly Australian Census, the Australian Bureau of Statistics asks people who speak a language other than English at home to state how well they speak English.

Respondents can choose from four options: “very well”, “well”, “not well”, or “not at all”. The categories “not well” and “not at all” are reported together.

In the 2016 Census, 4.9 million people reported speaking a language other than English at home.

Of those people, the number of people who reported they spoke English “not well” or “not at all” was 820,000.

Hanson was correct to say the number of respondents who ticked the “not well” or “not at all” categories has been rising – from 560,000 people in 2006, to 655,000 people in 2011 and 820,000 in 2016.

But of course, the overall Australian population has also grown over that time. So let’s look at the numbers as a proportion of the broader Australian population. On this measure, it amounts to a rise from 2.8% of all Australian residents in 2006 to 3.5% in 2016.

Over the same time, the percentage of bilingual residents who reported speaking English “not well” or “not at all” fell slightly, from 17.5% in 2006 to 16.6% in 2016.

That means within the bilingual population, there was an improvement in perceived English language skills between 2006 and 2016.

Hanson said there were now “over a million” people in Australia who “cannot speak English well or at all”. There are two potential problems with the calculations made to come to this conclusion.

Firstly: the calculation assumes the same rate of growth in the number of people who speak English “not well” or “not at all” between 2016 and 2021 as it was between 2011 and 2016.

The number of people with little or no English language capability is largely a function of the overall migrant intake. As our overall migrant intake has increased, the absolute number of new arrivals with little or no English language capability has also increased.

However, since the 1990s, our migration program has become increasingly selective and the English language requirements for permanent residency have risen.

Second, the projected growth rate suggests that not speaking English well is an unalterable characteristic, and that new entrants with little English capability simply add to the existing number.

This assumption doesn’t account for the likelihood that many recent immigrants who responded that they did not speak English well or at all in the 2016 Census will have improved their English (or their confidence, or both) by 2021 and will respond that they speak English “well” or “very well” then.

How accurate are the data?

The Census data provide us with a rough guide to English language proficiency, but it’s not a particularly valid or reliable measure.

That’s because the judgements made in the survey are subjective. There’s no definition around what speaking English “well” or “not well” means. One person may overestimate their English proficiency, while another person may underestimate theirs.

As noted by the Australian Bureau of Statistics:

… one respondent may consider that a response of ‘Well’ is appropriate if they can communicate well enough to do the shopping, while another respondent may consider such a response appropriate only for people who can hold a social conversation.

As such, these data should be interpreted with care.

Self-assessment can be a valid tool in determining language proficiency. But for that to be the case, the questions need to be much more detailed and sophisticated.

So while we can state that 820,000 Australians reported speaking English “not well” or “not at all” in the 2016 census, it’s not possible to determine what that means in terms of their actual ability to communicate in their everyday lives.

Most bilingual residents speak English ‘well’ or ‘very well’

The vast majority of bilingual Australian residents report speaking English “well” or “very well” – more than 4 million out of 4.9 million.

Evidence of a certain level of English language proficiency is a visa requirement for most permanent migrants, and many temporary migrants. The key exceptions are humanitarian and family reunion migrants, whose reasons for admission supersede the immediate language requirements.

New citizens are also subject to an Australian citizenship test, which is an implicit English language test, requiring a certain level of English language proficiency to pass.

The number of people in Australia with little or no English language capability depends not only on the number and mix of new migrants admitted, but the English language training provisions made available to those people when they arrive. – Ingrid Piller

Blind review

I agree with the verdict of this FactCheck. The sources used and conclusions drawn are correct. – Amanda Muller

The Conversation’s FactCheck unit was the first fact-checking team in Australia and one of the first worldwide to be accredited by the International Fact-Checking Network, an alliance of fact-checkers hosted at the Poynter Institute in the US. Read more here.

Have you seen a “fact” worth checking? The Conversation’s FactCheck asks academic experts to test claims and see how true they are. We then ask a second academic to review an anonymous copy of the article. You can request a check at checkit@theconversation.edu.au. Please include the statement you would like us to check, the date it was made, and a link if possible.