Government released the Terra Nova results for the 2008 academic year this week, and, if nothing else, they made the case for the need for education reform.

While there were some marginal improvements in the tests of language and reading, maths scores remained well below the US mean.

And the move of students from primary school to middle schools still causes worrying declines in performances for these age groups.

At first glance, this may seem to be caused by the trauma of the move. Or it might appear to be because, in education-speak, the primary and middle school curricula are not well aligned.

The likely truth is more mundane, and may say less about the performance of middle schools and much more about the perceptions of parents about public education.

Around 480 primary school students take the Terra Nova tests in each primary school year. But just 380 students from the first year of middle school. This is not due to a sudden aberration in the birth rate. It is because around 100 students from the Government primary schools move to the private sector when they are due to start middle school.

And in general – and this by no means is an absolute rule – the private schools get the higher performers from the Government school at this age.

To some degree, Government middle schools are in a chicken and egg situation. They cannot prove to parents inclined towards private schools that public schools provide as good an education because, having lost "the cream", their results are likely to decline. But improve they must, or they will never win the perception game.

That is why education reform is critical.

What is more worrying across the board is that even the best results from year groups in public schools (and these are only in the primary schools) barely pass the US median. So even our best year groups are at best mediocre.

Of course, there are individual schools who probably excel, although these results are still not being made public. And it is also true that in a school system that must take all age-eligible students, there will inevitably be poor students who will drag the mean down as well as good ones who will pull it up.

But overall, the system is below the US average, and the US school system is routinely ranked well down international rankings.

That is why education reform is critical.

For all these reasons, education officials prefer to track progress, as opposed to absolute results.

Even here, progress is patchy. In language and reading, primary schools show some progress as they move from one year to the next, before it drops off in the first year of middle school. There is then some improvement in the next two years of middle school which continues into high school without ever reaching the US mean of 50.

In maths, the picture is worse. There is very marginal improvement in primary schools, and then some slight improvements again after the first year of middle school. But it is nothing to write home about and it is worth noting that no year group in the past two years has ever reached 50 percent.

That's why education reform is critical.

And while there has been a good deal of talk about how the schools are going to be restructured, the only two key points to watch are these.

Teachers need to be qualified and held accountable, and the curriculum needs to be rigorous enough to raise standards. Nothing else, not cluster boards, not buildings, not the structure of the Ministry matters apart from that.

We need to find good teachers. We need the hold them accountable and back them when they deserve it and we need to let them teach.