OPINION: NAPLAN is a high-stakes test - schools are held publicly accountable for its results - and high-stakes tests are known to lead to unethical, unscrupulous, and sometimes corrupt behaviour.
Federal Education Minister Peter Garrett, perhaps mistakenly, recently overemphasised our students' low literacy levels, perhaps so he could promote score gains on NAPLAN. We are only just beginning to understand NAPLAN's huge impact on the way our teachers teach, how our school administrators lead, and how our policymakers shape the public conversation about educational achievement.
First, when students are tested with a high-stakes examination, some teachers will engage in ''test prep'', or teaching to the test. Some of this test prep, like teaching longer and working more effectively, may create legitimate gains in student learning. But other forms of test prep inflate standardised test scores and make student gains appear larger than they are. Teachers reallocate their content material: they will emphasise what is tested and de-emphasise what is not. This wouldn't be a problem if what teachers were teaching less of was not important, but it often is important, just not related to the test. Teachers also engage in coaching, where they use tests similar to those their students take to improve test-taking skills. This type of prep focuses on trivial aspects of the test, such as which multiple-choice answer students should select if they are unsure of the correct one.
Some school administrators will promote the same type of behaviour, but at managerial level. For example, school leaders will encourage, even mandate, the reallocation of teaching time from one non-tested subject to a tested subject. In the US, high-stakes tests have been around for many years and researchers have had time to study their effects. Tested subjects - such as mathematics and reading - get substantially more teaching time, and other subjects - such as art and music - were entirely cut from some school curriculums. Of course, some will argue that we shouldn't be concerned if our students are learning to the test: NAPLAN is what we want them to learn. This is false: on an alternative, reliable and valid test that measures similar content but in a completely different format, our students are likely to perform much worse than on NAPLAN.
But our education leaders will probably be tempted to spin wild stories about our score gains: everyone is dramatically improving, the gap between our rich and poor students is significantly narrowing, and everyone is above average! While some of these gains may be real, they may be grossly exaggerated for political purposes.
It's important to remember, though, that to blame individuals, including teachers or principals, is meaningless. The examples I list have nothing to do with personal character and all to do with context: any rational person, under pressure to demonstrate results, will have a strong motivation to engage in these types of behaviours.
What we are likely to see in the years to come is strong student improvement on NAPLAN. This will reflect the score inflation from test prep of the first few years. Then, our NAPLAN scores are likely to rest at an inflated plateau.
NAPLAN is a useful tool that can shed light on the most disadvantaged schools that need attention from social and educational policy. But, in the end, it gives us a very coarse average that many interpret as a reflection of the successes or failures of our schools. The underlying assumption, of course, is that if we make the schools better, student achievement will improve.
I am not arguing for ridding Australia of standardised testing: this would be fanciful and counter-productive. But we need an evaluation of the validity of NAPLAN score gains. Also, if instead of testing our entire population of students and schools, NAPLAN was administered yearly to a select, representative sample of students, we would still get relevant data, but the test would not be high-stakes, and so the problems I describe would be unlikely to occur.
Dr Leila Morsy is a lecturer in the school of education at UNSW. She has a doctorate in education from Harvard.
This piece was first published in the Sun Herald.