Whatever one’s view on the New Zealand education system, one point seems hard to deny. The achievement profile of our students shows an unusually long tail. While most of our students do very well, a significant proportion do very poorly, and the proportion is high by international standards. The make-up of our disengaged is skewed by race, income level and gender. The populations whose potential is most being stifled are also those with the highest birth rates, which suggests a mechanism by which the problem, if unaddressed, may worsen with time. This type of underachievement matters on every level. There is a human rights issue, a broader social cost, the significant economic disinvestment, and obvious implications for social cohesion. Nobody in their right mind believes this is an issue we should ignore. We’re failing.
The controversy arises when we ask who this ‘we’ is, and two predictable contenders have emerged. On the one hand is the point of view endorsed by the current government, and pushed strongly by leading civil servants like Treasury’s Gabriel Makhlouf and The Ministry of Education’s Lesley Longstone. It says, in essence, that part of the problem lies with the education system itself. We can and must do better as teachers to identify and rescue these learners. Longstone has gone as far as to say that we can’t claim a world class education system while this tail persists.
Against this, is the view most strongly advocated by teacher unions. It says, we’re actually doing a pretty good job, perhaps even a world class job, with all our students, and the failure belongs to the broader society, in that some students don’t bring to school the necessary skills upon which an education is built. This is sometimes summarised as the poverty explanation, although that’s slightly misleading.
It doesn’t take a rocket scientist to see that this debate is drawn along classically political lines. If you lean to the right, the heroic narrative is one of individuals able to rise above their circumstances, and turn their lives around. The Makhlouf/Longstone story is immediately compelling. If you lean to the left, the heroic narrative speaks of societies defeating the evil of inequality, and so freeing its citizens. If this is your favoured story, the tale told by the unions makes intuitive sense.
The optimist in me believes that if we look carefully at the data, we can work out which argument is more compelling. The pessimist in me notes that so far, nobody has.
Here’s my sketch of the necessary investigation. Both Makhlouf and Longstone have pointed out that other countries with similar levels of poverty don’t necessarily have the same level of underachievement, and both have publicly drawn the conclusion that therefore there’s more to this than poverty. That, as far as it goes, is probably correct. Both have then taken a further step, and suggested that given there’s more to this than poverty, we need to look at how the education system is failing these children. And that next step, logically speaking, need not follow. What’s more, I suspect that it doesn’t follow in fact either. I could be wrong, but here’s how we could find out:
As a secondary school teacher, I have around about 120 hours contact with any given class over a year. If those classes, on average, have 25 students in them, then contact time per student over the year averages out at a shade under five hours. That’s the time available for one on one interaction, be it feedback on a piece of work, explaining a problem, responding to a joke, or wishing them luck for the weekend’s sports game. In reality, it’ll be less than that. The same student, in the course of a year, will be awake for a little under 6,000 hours. So, I get less than 0.1% of their waking time to make a difference. Primary school teachers get more, maybe as high as 0.5%, or one in every 200 hours. We do have an impact, we do make a difference, but immediately it should be clear that so too do a whole heap of other people and influences. Formal education represents just one of a number of variables. What’s more, if the neuroscientists who insist that the first three years of development are the vital ones are right, educators have no presence whatsoever during the vital phase where the intellectual foundations are laid.
So there’s a flaw in the Makhlouf/Longstone case. Poverty needn’t just mean lack of financial resources. In educational terms, poverty might refer to a deficit in the dominant educational environment, which is clearly the non-school environment. Income levels are almost certainly a pertinent variable, but so too are parental education, trust of institutions, dominant language in the home, drug habits, accessibility of role models, exposure to violence, involvement with the health system, access to books in the home, etc. Because there are so many factors in play, showing that one factor (poverty) doesn’t account for all the variance doesn’t automatically suggest one other factor (teaching) is part of the problem. It may well be that we are doing an excellent job within the formal education system, but the effect is being swamped by other, more powerful factors. There’s clearly a difference between a financially poor household and an educationally poor one.
One reason to suspect this is in fact the case, was proposed recently by Wellington College principal, Roger Moses, who noted that intuitively it seems odd that the same teachers who do remarkably well with 80%, somehow are doing a lousy job with the other twenty percent. That’s not an argument in itself, but it’s a good reason to interrogate the hypothesis. The decile distribution of recently released primary school data, is another.
In science, the way we interrogate an hypothesis is to test it. We ask, what should I expect to see, if this hypothesis is correct, and then go and have a look. If we apply that approach to the Makhouf/Longstone hypothesis, something interesting happens. If they’re right, and part of the problem is the education system, then we should expect the problem to be responsive to time spent in that system. (If my hypothesis is that one of the reasons my carpet is fading is exposure to the sun, then I should expect that the more exposure there is, the greater the fading). If however, the problem is primarily background poverty (in the broader sense) we should expect to see that the problem already exists when the students enter the compulsory education sector, and isn’t greatly magnified by teacher exposure. In other words, if achievement distribution at fifteen is much more unequal than it is at five, then the schools absolutely need to look at how this is occurring. If, however, the gap is static, or even closes over that time, then schools aren’t the primary problem.
At this point I must note that it would be logically fallacious to suggest that just because schools aren’t causing the problem, they can’t be used to help solve it. There is no logically necessary relationship between a problem’s cause and its solution. The greater point, rather, is that it’s much harder to solve problems when we don’t understand their causes. And it’s unhelpful to pretend we understand the causes, when we don’t.
We have the tools to do this analysis, indeed the much decried National Standards, should they ever achieve a degree of reliability, are perfect. Even without them, schools already have good data on student achievement. So, here’s my question. To what extent do inequalities increase during the school years, and to what extent are they already in place at entry level? If we can’t answer that, we don’t get to have an opinion on the role education is playing in failing our most vulnerable. Time for those with the resources behind them to step up to the plate.