Monday, 14 March 2016

RSAS Up: The Commonwealth's Remote School Attendance Strategy

The Department of Prime Minister and Cabinet (PMC) published, on 4 March 2016, an interim progress report dated October 2015 on the Remote School Attendance Strategy (RSAS), the Commonwealth’s flagship initiative in relation to remote education.

The Commonwealth also supports a number of smaller pilot initiatives such as the SEAM program which operates in a number of remote and urban NT (and previously Qld) communities and links school attendance to parents’ welfare payments, and the Learning on Country Program which operates in four Arnhem Land communities.

The SEAM program was assessed by the ANAO in a 2014 audit which found ‘mixed’ results, and the program was evaluated in a May 2014 report which appears very comprehensive, but largely equivocal in terms of its results. The Learning on Country program was evaluated in a report dated May 2015; refer to Minister Scullion’s media release dated 15 November 2015 reporting a qualified yet positive evaluation of the pilot. In addition, one of the five components of the Indigenous Advancement Strategy (IAS) relates to Children and Schooling, and presumably this component funds initiatives proposed and implemented by third parties.

RSAS commenced in 2014 and involves the engagement of around 400 local school attendance supervisors and officers to work with students, families, and school communities, involving 77 remote schools. The program was rolled out in two phases in 2014, across five jurisdictions, and was funded for two calendar years, 2014 and 2015 at a cost of $46.5m. In September 2015, the Minister for Indigenous Affairs announced that the ‘successful’ program was being extended for a further three years to the end of 2018 at a further cost of around $80m.

Extraordinarily poor educational outcomes in remote Australia have been an issue of concern for at least two decades. In our 2007 book Beyond Humbug, Neil Westbury and I identified education as an example of governmental incapacity to implement effective delivery of essential services. We cited 2005 Productivity Commission data which indicated that in remote areas of the NT in 1999, only 3 to 4 percent of Indigenous students achieved the national reading benchmarks.

The most recent NAPLAN report includes data which indicates that current Year 3 reading achievement of Indigenous students in very remote regions of the NT is that 27.4 percent are at or above the benchmarks, and 71.2 percent are below. The comparable national figures for very remote regions are 52.3 percent below the benchmark and 46.6 percent are above (refer Table 3.R6).

For Year 9 students, in very remote NT, 84.6 percent are below national accepted benchmarks, while in very remote regions nationally 69.7 percent are below national benchmarks (refer Table 9.R6).

Clearly there is a strong policy rationale for additional government action. Two in three remote Indigenous students are not achieving a year nine reading standard, with the concomitant lifelong consequences for employment, income-earning capacity, and the substantial personal and social opportunity costs which that entails.

The issue then is what is the appropriate and/or most effective policy response? And which level of Government should drive that response?

Education is primarily a state and territory responsibility, but is an area where the Commonwealth has longstanding interests, normally advanced through the provision of funding to public and private education providers. The Commonwealth has long-standing mechanisms for using the power of the purse to incentivise states and territories to advance Commonwealth priorities.

For the Commonwealth to directly intervene in one aspect of the education system, Indigenous student attendance, as opposed to the other areas which influence indigenous educational outcomes (teacher education, teacher quality, curriculum, capital works, facilities, overall resourcing levels, etc) and to do this only in a subset of all remote schools is, in my view, the most risky and potentially most problematic approach to addressing the issue of poor educational outcomes.

Inadequate investment in all aspects of education by the states and territories, both financial and intellectual, has clearly been a longstanding issue in remote education provision. Unless the Commonwealth is absolutely certain that there is only one issue in remote education, and that issue is poor attendance, then the Commonwealth’s actions are likely to lead to sub-optimal outcomes, and most importantly, will delay the day when effective policy responses will be brought to bear.

In particular, the states and territory have little incentive to take responsibility for outcomes, because in political terms, the success or failure of remote schooling will be seen to be the responsibility of the Commonwealth.

School attendance is an issue the Commonwealth has traditionally left to the states and territories. School attendance is certainly a function of a complex array of factors outside of education, some in the Commonwealth’s domain like welfare policy, but most in the domain of state and local governments. More importantly, school attendance is also a key component in the system comprising a set of complex and inter-related administrative, intellectual and professional activities, which together comprise the schooling system. This system is fundamentally the responsibility of the relevant state and territory education departments.

So I would argue that any policy response to poor educational outcomes should, in the absence of a strong countervailing rationale, be implemented by the states and territories.

As for the design of the most appropriate intervention, this is a huge topic, and I do not claim particular expertise. My own instincts however are that ensuring that remote schools have access to experienced and high quality teachers (with a capacity to engage with their school community as well as running effective education programs) and an effective curriculum are key. Noel Pearson and Chris Sarra have both written persuasively on these topics and have effectively laid out a policy roadmap for governments to consider. If quality teaches and curriculum is the key, then while attendance is important, it becomes essentially a second order issue (insofar as getting the right teachers and curriculum will assist in maintaining attendance rates at acceptable levels). Or to put it another way, high attendance without quality teaching involving high expectations and a quality curriculum will achieve very little.

For better or worse, the Commonwealth has decided to focus its additional intervention virtually solely on school attendance through RSAS and to a lesser extent SEAM. PMC is to be congratulated for releasing the interim progress report on RSAS and for commissioning the evaluative and analytical work which underpins it.

The report includes both quantitative and qualitative components, and concludes that the program has had a positive impact on school attendance overall with an increase in attendance over the first year in a majority of schools – in Queensland and the NT, 72.5 percent of schools (29 out of 40) had an increase, and across the NT, the average number of students on any one day in term three of 2014 was 13 percent higher than in term three 2013.The qualitative evidence was much less clear cut, identifying a range of reasons for poor attendance cited by departmental staff in their weekly reports, but also noting significant issues with recruitment and retention of staff to deliver the program.

The report raises at least four significant issues which are worthy of comment.

In relation to state and territory governments the report notes that they ‘are simultaneously operating their own school attendance programs and strategies which RSAS aims to complement’ (page 3). However the qualitative report which is based on the traffic light reports provided by RSAS staff each week has very little to stay about the inter-relationship with these programs. Moreover, it is apparent (but not explained ) that at least two Governments, South Australia and Western Australia were not prepared to allow data from their states to be utilised in full in the report’s analysis. This is in my view a significant issue as it goes directly to the potential of the program to influence outcomes more generally in the state education systems.

A second comment concerns the data for particular schools. What is immediately apparent from an examination of the tables (and is glossed over by a focus on the percentage change, positive or negative) is that there is a large variability in the absolute attendance levels amongst the various schools, and indeed a number of schools which are improving in percentage terms are nevertheless subject to extremely low attendance levels, particularly in the NT. Thus Table 1.3 on page 6 indicates that 19 of the 29 schools in the RSAS program in the NT had absolute attendance levels in 2014 of less than 60 percent, and 7 schools had attendance levels below 50 percent.

Moreover, the review analysis is focussed entirely on average attendance levels at each school. Yet learning is incremental and requires sustained skills acquisition. Learning deficits quickly emerge if there are gaps in attendance. So it is important to understand whether the non-attendance is confined to a particular cohort within each school, or is more widely shared. Fifty percent attendance might mean 50 percent of students attend every day and fifty never attend, or it may mean that all students miss 50 percent of the school year. The latter outcome has much worse educational consequences than the former, but these differences have not been addressed in the review’s analysis.

Finally, the qualitative analysis is based entirely on the reports of the Commonwealth officials working in each location. This is an inherently limited information base, and merely compiles and analyses the reasons listed for attendance changes by RSAS staff. There is a place for qualitative analysis, but it needs to test particular hypotheses and should engage with actual stakeholders (students, parents, teachers, community members) to have any chance of drawing insightful conclusions which policy makers might then use in devising policy or adjustments to policy.


The Commonwealth’s current policy on remote school attendance appears to be fundamentally flawed. It bears all the hallmarks of a policy initiative designed to be seen to be doing something, yet runs the risk that it will actually allow the states and territories off the hook . There are clearly interactions with the operation of SEAM in the NT (administered by PMC and DHS), but neither program’s design logic appears to recognise the existence of the other. RSAS operates in a limited number of remote locations, and thus will only ever have a partial impact. A more effective alternative would have been to allocate the funds to the relevant education departments utilising an incentive structure which rewards not merely improved attendance (an output), but ideally improved NAPLAN scores (an outcome), leaving the methods to be employed to the education experts.

Of course, RSAS clearly has political benefits for a politician seeking to win votes, particularly in the NT, in that it sets out to create employment for local community members across 29 key school communities. Interestingly, the more structural impediments within communities for improved school attendance appear to make these jobs quite difficult to carry out, leading to high turnover and poor retention. Nevertheless, the potential political motivation is easy to discern. Moreover, naming a program a strategy doesn’t mean that there is a strategy. The Commonwealth has not published a comprehensive policy based rationale for the program.

There would be merit in developing and publishing a strategic plan (or mini white paper) on the overall Commonwealth’s strategy for achieving improved educational outcomes in remote Australia. Such a plan would ensure that a comprehensive and coherent program logic would be devised, and would canvass how best to harness the resources and expertise of the states and territories, and thus lay out a comprehensive rationale for the Commonwealth’s involvement.

If the states and NT were not prepared to cooperate, the Commonwealth should then canvass options for the Commonwealth to take over the whole school education system in remote Australia from start to finish rather than inject random interference as at present.

In the absence of such a strategic rationale for the Commonwealth’s involvement in remote education, interested citizens can be forgiven for seeing RSAS as merely another instance of politics subverting good policy. It will most likely end up on the scrap heap of failed policies in Indigenous affairs, with Indigenous citizens wearing the reputational damage of yet another policy fiasco, taxpayers being $125m worse off, and yet another generation of remote citizens reaching adulthood without the literacy and numeracy skills which will allow them to fully participate in our nation’s future.

When Noel Pearson referred to a crisis in Indigenous policy, he was talking about more than education and schooling. Yet it is apparent that the crisis runs deep in remote education and is embedded not so much in the communities, but in the structures of government itself. Part of the reason we have as a nation found it so difficult to address indigenous disadvantage is that we have been looking for solutions in the wrong places.