WE DON'T NEED NO EDUCATION? (Part 1)

It's time to stop dithering and make Australian schools great again

Commonwealth, State and Territory education ministers met on 6 July (communique). There were some useful outcomes. Though I’m not an expert on teacher training, requiring universities to instruct prospective teachers on how to manage a classroom seems to me to be a valuable step forward.

What struck my interest though, having researched school and student performance previously, is the update on the Review to Inform a Better and Fairer Education System. Education ministers established an expert panel in December 2022 to inform the next Commonwealth-State agreement on how to make schooling better and fairer.

The Review is an opportunity to drive real and measurable improvements for all students, particularly those most at risk of falling behind. The Review will build on the work of the Productivity Commission and their review of the NSRA, released on 20 January 2023. This review highlighted key areas where reform is needed, including lifting student outcomes, reducing differences in outcomes across students, and supporting student wellbeing.

Source: education.gov.au

Italics are mine. Two comments.

First, the wording is not ‘The Review will deliver’. No, the Review is merely an opportunity (i.e. it has a non-zero chance) to drive (i.e. let governments pick up recommendations they were inclined to pursue anyway) improvements.

Second, the Review follows a 6-month review by the Productivity Commission, yet this Review to be informed by the earlier Productivity Commission review is to take two years.

While I love an expert panel as much as the next former public servant, I’m not sure why we need to wait two years to start the radical changes on the ground needed to reverse the collapse in student outcomes.

The update the Review Panel provided to education ministers noted they had done a lot of things: meetings have been held with jurisdictions and stakeholders across the country, including visits to some schools; views have been received from over 24,000 teachers, parents and students in response to a survey. And on 5 July 2023, the Panel released a Consultation Paper. (It’s a decent read if you have the interest.)

The wisdom passed down from Yes, Minister is that there are only three reasons to hold a review: because you don’t know what to do; to build consensus; or to delay action.

My sincere hope is that the purpose of this review is to build consensus.

Having worked as the Commonwealth’s lead representative in a different area of public policy, my guess is that the Commonwealth already knows what its dream-state outcomes are, that State and Territory bureaucrats probably agree with 80% of those ideas (although not necessarily the same 80% in each State), and the main task is to create time to see how much those state bureaucrats can get their ministers across the line on.

Education reforms are especially tricky to land at a Commonwealth-State level because:

• the Australian Education Union is powerful in stopping things, and

• the Commonwealth only holds convening power and the purse-strings over minor funding at the margin.

Part 1 of this post argues that the problems in schooling are clear; Part 2 will argue that at least a few of the solutions are clear too.

MEASURING THE PROBLEM

The worsening performance of Australia’s schooling system is well-documented.

The gold standard for international comparisons of academic attainment in schools is the OECD Programme for International Student Assessment (PISA). It regulates carefully the conduct of aptitude tests across countries. e.g. schools need to be chosen at random with only narrow grounds for excluding special education schools. In Australia it involves about 15,000 children at age 15.

So let’s talk about the ‘real and measurable’ degradation of the education system.

PISA scores are compared to a benchmark level of student achievement of 500 that was the average across the OECD in 2000. At that time, Australia placed relatively well, but is now middle of the road. Student achievement in mathematics has plummeted particularly sharply.

AUSTRALIAN SCHOOL ACHIEVEMENT HAS FALLEN, SHARPLY
OECD average in 2000 = 500

Source: OECD

Another way to cut the data is by comparisons with other countries at a point in time (table). Our children are faring worse in terms of average student achievement than those in the UK, Canada or New Zealand.

Shading represents those countries with a greater than average share of top performers.

Again, the weak performance of Australian students on mathematics stands out today, on both an average and top-performer basis, whereas our students were performing well two decades ago.

WELL, WE AIN’T ESTONIA, THAT’S FOR SURE
PISA scores in 2018

Source: OECD.

Is this just smartphones destroying a generation and the rise of social media that distracts kids from their studies?

An easy riposte is, no, because all countries on the list have seen essentially the same changes in social media access and many have sustained high scholastic performance.

But the more worrying issue is that Australian schooling outcomes began declining much earlier.

The downward trend in student attainment evident in the PISA surveys continues a trend in earlier domestic surveys. Andrew Leigh and Chris Ryan analysed earlier surveys taken of students at age 14 years using consistent survey instruments between 1975 and 1998. They find a statistically signficant fall in achievement. Australian students as a group have never again scaled the lofty heights they achieved in 1975.

THE DECLINE IN LITERACY AND NUMERACY BEGAN DECADES AGO
Academic achievement of 14 or 15 year old students
OECD average in 2000 = 500

Source: 1hand calculations based on OECD PISA and Leigh and Ryan (2011)

(Wonknote: 1975 to 1998 was a temporally consistent set of literacy and numeracy tests and scores are based on common questions across the surveys. We link those historical results to PISA scores, taking accounts of differences in standard deviations, by assuming 1998 domestic scores are the same as Australian PISA scores in 2000. Regardless of any obfuscation required to explain the joining of the series, the falls within consistent surveys from 1975 to 1998 and from 2000 to 2018 are both clear. Calculations below assume a normal distribution.)

To stretch the statistics as far as we can: the level of literacy and numeracy obtained by the average student in 1975 is today only exceeded by 30 per cent of students, with 70 per cent of students attaining weaker competence.

That is, as a community, we have gone backwards.

We are teaching our children less well today than we did 50 years ago.

We are teaching our children less well than we were taught.

You are teaching your child less well than you were taught. On average.

A moment to pause.

Is this because more students ill-suited to academics are staying at school too long and hence are more likely to appear in test samples? No, all of these surveys were undertaken at ages that Australian children were required by law to attend school at that time.

Is this because of difficulty teaching the growing population of migrant children from non-English speaking backgrounds? No, the OECD reports that in 2018 29% of Australian migrant children achieved in the top 25% of all Australian students for literacy. That is, the contrary hypotheses that the average migrant child was being taught more effectively in their country of origin before migrating than Australian-born children were being taught in local schools, or that they attach greater importance to education because of the values of their parents, or both, are worth exploring.

No this is not due to noise in the data.

However you want to cut the data.

We are teaching our children less well than we were taught.

BUT WHO CARES WHAT THE OECD SAYS?

In Australia, those 15,000 students that completed the PISA each wave have then been tracked to age 25, and it turns out that PISA scores at age 15 are the single most-powerful predictor of whether a child goes on to a university education.

If you want to make schooling better and improve post-school opportunities for Australian children, it has to start with raising literacy, numeracy and science skills in the early years.

Another way to understand why these declining results matter is to recognise that far fewer people from disadvantaged backgrounds attend university.

Differences in literacy, numeracy and science as measured by the PISA explain 18% of the gap between university attendance of urban and regional populations, 29% of the gap between Indigenous and non-Indigenous populations, and 52% of the gap between those growing up in the lowest income quartile compared with other groups. These results are calculated after controlling for a long list of other contributors: gender, public versus private schooling, which state you live in and even the average socio-economic status of people in the suburb you grew up in, among other variables.

If you want to make schooling fairer, it has to start with raising literacy, numeracy and science skills in the early years.

LITERACY AND NUMERACY AT AGE 15 EXPLAINS WHO GOES TO UNI
Share of the gap in university participation explained by PISA scores

Source: Productivity Commission, The Demand Driven University System: A Mixed Report Card

SUMMING UP

The performance of Australia’s schooling system has declined steadily since 1975. Where we used to be a high performer, now we are a mid-performer overall and weak on mathematics.

The education system needs more than a nip here, a tuck there.

And it needs action now. Changes shouldn’t be put on hold while waiting 2 years to hear from a review panel.

More to come in Part 2 on what some of the solutions are.