Australian education ministers are being duplicitous on school league tables. They claim to be opposed to ‘simplistic’ league tables but are providing the information to enable their publication.
League tables are inevitable because school results are to be published on a centralized national website, as well as on several state and territory websites. It will be a simple matter for anyone to construct school rankings from them.
The evidence against publishing school results and league tables is overwhelming.
First, league tables restrict student learning because they narrow the curriculum and teaching. Students receive a less rounded education.
Overseas evidence shows that schools direct more resources into the tested subjects of literacy and maths while untested subjects such as science, history, social studies, languages, arts and music, physical education and health receive much less time. There is also less teaching of more complex thinking and writing skills.
League tables turn classrooms into test preparation factories. Weeks and months are devoted to test preparation instead of deep learning.
Second, league tables punish low ranking schools, their teachers and students with an annual public ‘tar and feathering’. Students who are publicly humiliated are unlikely to respond positively in their future learning.
League tables serve as a job guide for teachers to apply to highly ranked schools with fewer learning and behavioural problems. This means that low ranked schools often end up with the least qualified, least experienced teachers.
Third, league tables mislead about school quality because they reflect factors outside the control of schools and teachers. For example, school results are significantly determined by the socio-economic background of school communities. League table rankings are more a measure of the family background of a school’s students than the quality of its teaching.
Also, if a higher proportion of families engage private tutoring a school will receive a boost to its measured performance and league table ranking even though there was no change in teaching effectiveness.
School results are often manipulated in response to competitive pressure to improve rankings. Overseas experience shows that many schools rig their results by poaching high achieving students from other schools, not admitting low achieving students or suspending them on test days, helping students with answers or changing answers.
The Texas Education Authority has just introduced a scheme which allows schools to report students as passing tests even when they fail. Schools will be able to do so if a complex formula predicts that students who fail will pass in a future year. It will allow hundreds of schools to boost their performance.
Fourth, league tables tend to increase social segregation in schools. This exacerbates inequity in education as student outcomes tend to be lower in schools with high proportions of students from socio-economically disadvantaged families.
Social segregation in schools also undermines understanding and tolerance between different social groups in communities and workplaces.
Julia Gillard wants to report school results so that “parents drive change” for school improvement through greater competition. But, as Professor of Economics at the University of Chicago and the co-author of the best-selling Freakonomics, Steven Levitt, says: “the theory sounds great, but evidence confirming it has been hard to find”. Many sophisticated studies support this conclusion.
Gillard says that so-called like school comparisons will provide contextual information that obviates the harm of simplistic league tables. However, like school comparisons are partial league tables and they incur the same problems as full league tables.
They narrow student learning in the same way. They provide the same incentives for schools to rig their results. The lowest ranked schools in each like school group will be pilloried and humiliated.
Partial and full league tables discourage collaboration and co-operation between schools. Schools will be reluctant to share successful practices with other schools if it means those schools could leapfrog them in ranking.
The existing ways of grouping like schools in Australia, and Gillard’s favoured New York model, fail to compare like with like on several counts. The measures of the socio-economic composition of schools are unreliable. For example, there may be larger differences between schools at the higher and lower boundaries of each group than between schools clustered either side of group boundaries.
Another problem is that the comparisons generally fail to distinguish different ethnic populations of schools which have a bearing on school results.
Australia achieves amongst the best student outcomes in the world without reporting school results and league tables. The highest achieving countries such as Finland and Korea do not rank schools. Student achievement in countries that rank schools is well below Australia’s. Gillard’s new reporting regime is likely to send Australia backwards.
Australian governments should subject school performance reporting and league tables to the transparency test they want to apply to schools. They should initiate a parliamentary or independent inquiry into the potential harms and benefits.
If governments fail to heed this, the only recourse for parents and teachers is to boycott the national tests next year.
Trevor Cobbold
This article was published in The Canberra Times on July 24 2009