Deciles a blunt instrument for school choice

The Dominion Post
28 February, 2019

Comparing schools is complex at best – and a nightmare at worst. 

Every school has a different cohort of students, each with their own unique background. But like one's starting position in a 100m sprint influences where you finish, one's background influences academic and school performance.

Some of the students who started college this month would have a head start – these students would have already spent thousands of hours reading books, playing music and participating in sports. 

Other students are already behind. They may struggle with basic reading and writing, have never had a single music lesson, or played a game of soccer, rugby or netball.

Disparate cohorts make it difficult to compare schools, particularly of different deciles. High decile schools have more students with a head start, while more students in low decile schools need help catching up. 

Comparing different schools in New Zealand is a routine but essential task for the Ministry of Education, teachers and parents. However, this complex process is holding back New Zealand's education system – and failing our children.

School choice and competition is the third key issue under the Tomorrow's Schools education reform. The reforms highlight and recommend solutions to growing segregation in New Zealand schools, or 'white flight'. This issue is a result of parents seeking the 'best' school for their children. However, when parents lack information on school quality, they tend to resort to using decile as a proxy for school quality – further aggravating socioeconomic segregation in schools.

This is reflected in the Ministry of Education's data: in the decade following the year 2000, the number of European students attending low decile schools halved, falling from 60,000 to 30,000. 

To improve school choice and competition, many countries have adopted value-added models to measure both student progress and school quality. Unlike typical assessment methods such as NCEA, which only measure where the student is at the end of the year (league table results), value-added models measure progress from the beginning to the end of the year (and progress over several years). 

By using a student's previous results (ie their results at the beginning of the year), value-added models can control for the different background characteristics of each student. The benefit is a system that can fairly compare schools with different cohorts. The greater the (academic) value a school adds to its students, the higher the quality/performance of the school. 

To be clear, value-added measures are not the only factor to be used to evaluate school performance. Interviews with the school's principal and teachers also contribute. 

Value-added models are not new. Since 1992, when the first such model was implemented in Tennessee in the US, they have grown in popularity all around the world. More than 30 US states and districts, the United Kingdom, Hong Kong and Australia have operationally used value-added models. 

In all these jurisdictions, value-added models have been used to give parents valuable information on how their local schools are performing. In both Tennessee and New South Wales, parents have access to government websites where they can see how one school directly compares with another. The information from the value-added models is translated into useful graphs and statistics that parents can use to make an informed and fair decision on school choice. Without this information, parents are left with making decisions in the dark.

Successful international value-added models can provide a road map for education policy in New Zealand.

Parents in Tennessee and New South Wales don't make decisions in the dark. They have access to specific information on how their schools are performing at the click of a mouse. Parents in New Zealand too deserve quality information on school performance so they can make informed choices.

Better information on school performance would release parents from relying on deciles as a proxy for school quality. The demand for high-performing low decile schools would subsequently increase, while the demand for low-performing high decile schools would decrease. 

For New Zealand to have any shot at reducing segregation in schools, better transparency in our education is the best option.

Non-biased information on school quality performance is also valuable for principals and the Ministry of Education.

Of course, principals know how well their school is doing, but additional information on how other schools are performing would help principals collaborate with those schools. 

By identifying the top-performing schools in New Zealand, the Ministry of Education could study what they are doing well and use this to determine best practices in New Zealand schools. Conversely, the ministry could identify and give the worse-performing schools appropriate support. 

New Zealand is facing an education crisis of a generation, and without better policy we are consigned to seeking solutions in the dark.

Stay in the loop: Subscribe to updates