Some evidence from UK suggests that publishing league tables improves schools

Opinion: Temporary abolition of tables in Wales seen to have increased inequality

Why in Ireland do we cling to the belief that comparing schools and making them academically accountable is a bad idea? Photograph: PA
Why in Ireland do we cling to the belief that comparing schools and making them academically accountable is a bad idea? Photograph: PA

The Irish Times feeder school list has been an extraordinarily popular feature over the past 10 years.

While the information is of much interest to parents, it is not officially released. Rather it is compiled by this newspaper from a variety of sources such as third-level institutions and the State Examinations Commission.

Of students from a particular school who started college this autumn, there is no way of filtering those who sat the Leaving Cert this year from those who sat the exam in previous years. So the information we are publishing today does not tell you how many Leaving Cert students from this year started in third level in Ireland this year, rather how many students who ever attended each school went to third level this year.

There would be no need to compile lists like the one today if the information was released officially, but we are told that information about exam results or college progression for individual schools would do far more harm than good if it were released. Indeed, the Minister for Education can actively prevent the release of data that could be used to compile comparisons of the academic achievement of students in schools. He has this power under the Education Act and, according to the department, successive ministers have considered the potential benefits and disadvantages of the compilation and publication of league tables based on examination data, and on that basis they have exercised their discretion not to release it.

READ MORE


Improving performance
But what if the ministers have been wrong? What if official school league tables have been shown to improve the performance of schools and lessen educational inequality? What if school league tables are good for education?

Well three years ago a research paper with precisely those findings made waves across the water in the UK.

England and Wales have very similar education systems. Between 1992 and 2001, both English and Welsh systems published annual school performance tables, based on GCSE (Junior Cert level) exam results. But then, in 2001, the Welsh parliament voted to stop.

Here was a natural experiment between two identical systems, one of which now lacked a key component of accountability: the official school league table. Researchers at Bristol university, led by Prof Simon Burgess, decided to look at what happened next.

The result, according to their findings, was “systematic, significant and robust” evidence that abolishing school league tables reduced the academic effectiveness of Welsh schools.

Results fell dramatically. Welsh GCSE performance levels fell by almost two grades per student, per year. That’s a student who would have managed a B in a subject scraping through with a D. The effect was spread across the different subjects, but two full grades worth of a decrease is a huge effect, even when spread across five or six exams.

Until 2001, the performance of students in England and Wales was very similar. After league tables were abolished the performance of Welsh students deteriorated. If , rather than abolishing the league tables, the Welsh authorities had opted to increase class sizes by one-third, this would have had a similar effect, according to the researchers. The drop in standards was also confirmed by the Programme for International Student Assessment.

Interestingly, this negative effect was concentrated in the bottom 75 per cent of schools. The top 25 per cent were unaffected by the abolition of league tables. Their results remained as high as ever and in line with their English counterparts. The abolition of the league tables had actually increased educational inequality.

Another charge often levelled at league tables is that they tell parents where the more advantaged students are and contribute indirectly to a level of demand on places that enables schools to cream off the more desirable students.

Burgess’s research, by contrast, found that the Welsh policy change had no systematic significant impact on either sorting by ability or by socioeconomic status. So why do we in Ireland cling to this belief that comparing schools and making them academically accountable is a bad thing?

It is true that other research on the issue has produced more mixed findings, but the research conducted by Burgess and his team is quite uncommon and deserves attention.

Unusually for this sort of area they had a test group (the Welsh students) and a control group (the English students.) The systems and students were identical apart from this one factor. There were no other major changes to the Welsh system. It is rare to be able to measure the effect of a school league table on an education system in as pure a way as the Bristol researchers did.

Perhaps unsurprisingly, Welsh authorities reintroduced comparative schools data – not league tables they insisted – in 2011.

We must ask why we aren’t even having a debate about the publication of performance tables in Ireland.


Beholden to schools
As it stands, parents are almost entirely beholden to schools when it comes to the amount and type of information that is available to them. Inspection reports are, of course, published and available online, but comprehensive though they are, they are not an accessible form of information for most parents. Comparing one school with another by means of an inspection report is well-nigh impossible.

Exams data is not the whole picture but it's an important part of it. Parents know that, and until they are given more transparent information about the schools their children attend, lists like The Irish Times feeder schools will continue filling the vacuum.


Gráinne Faller writes on education for The Irish Times. The research cited above can be found at iti.ms/192XKdj