To follow the public debate on why some children prosper in school and others falter, you’d think it was all down to teachers.
The media – from the New York Times to the Sydney Morning Herald – as well as public figures and politicians are quick to blame any student failures on inadequate teaching. One journalist wrote recently: “…anyone with an ounce of brains knows what must be done. It’s time to move from identifying failing schools to identifying failing teachers.”
The problem with confident statements like these is that many factors affect a child’s school progress – genetic endowment, classroom peers, to name just two.
So sorting out how much of the differences among students is due to teacher quality and how much to other factors has challenged researchers for a long time – and all the methods invented so far have their limitations.
But they can be useful, particularly in putting the debate about teacher performance into perspective.
A twin solution
One way we have chosen to tackle the challenge is by analysing the early literacy performances of around 700 pairs of twins in NSW and Colorado schools.
Twins can help us because members of some pairs are in the same class and others are in separate classes. Although we know that members of twin pairs, especially monozygotic twins (also known as identical twins), score very similarly to each other, if teachers have the kind of impact some think they do, we should expect to see quite substantial differences between twins when they have separate teachers. And when they share a teacher, have much the same results.
Here are the pair correlations that we found for combined reading and spelling scores for the first three years of school:
The average overall difference between same- and different-class twins was about .08, which translates into about 8% of the variability among the children being due to the classroom they are in.
This is well short of the 40% sometimes cited and in line with some of the higher quality evidence from other research. We also know that classroom factors other than the teacher, sometimes called “classroom climate”, can affect the average class score, so the actual contribution of teacher quality is probably less than 8%.
Education researchers hold that teacher quality will show up most clearly when a child has a “bad” teacher several years in a row. In our project there was no support for that idea.
In a separate Scandinavian sample, we included many twin pairs in the same classroom and whose classes remained the same, with the same teacher, in the first and second grades. If teacher quality had the broad-brush effect many claim, we’d expect the twins to be even more similar to each other at the end of the second grade, compared to the first. But they were not.
The within–pairs correlations actually dropped a little – in other words, no sign of a cumulative teacher effect.
This study also shows us a higher correlation among monozygotic (identical) twins, compared to dizygotic (non-identical) ones. This shows us that genetics have a substantial effect on early literacy.
Genetics and literacy
Though we hasten to add that it is impossible to say in individual cases of reading difficulties whether genes or other factors are the cause, or are a combination of both.
We also need to emphasise that even in cases where difficulties are due to genetic factors there are very good reasons to be optimistic that well-focused intervention using proven techniques can alleviate them.
The kind of research we have reviewed, including our own, typically utilises large numbers of children, and large numbers of teachers and schools as well. As such, it can fail to detect individual acts of teaching and consequences of school policy that elevates a child beyond expectation or leaves another child foundering.
The research also does not mean that serious social disadvantage does not have its consequences. Indeed, research into the role of genes and the environment in academic and cognitive performance has shown that at low SES levels genetic influence is muted, as if the burdens of disadvantage prevent children from performing to their potential.
But the arena in which policy decisions are made, like the management of “bad” teachers, is the normal functioning of normal schools across large educational jurisdictions – this is where our data is useful.
Good and even
Our conclusion, then, is that teachers are doing a much more even job in fostering early literacy than many would have us believe. But is it a good job?
In the 2011 Progress in International Reading Literacy Study (PIRLS) of Grade 4 students, Australia ranked 27/50 in mean score, statistically indistinguishable for Bulgaria, New Zealand, Slovenia, Austria, Lithuania and Poland, above France, Norway, Belgium and Spain and the remaining participating countries, but below other English-speaking countries such as the US, Canada, Ireland and the UK.
In contrast, in less recent data on 15-year-olds from the 2009 Program for International Student Assessment (PISA), Australia ranked sixth among 44 OECD countries, on a level indistinguishable from New Zealand, The Netherlands, and Japan, and above other countries such as Sweden, Norway, Germany, Ireland, Denmark, the UK, and the USA.
It is not clear how to reconcile these rather disparate results. It could be simply a matter of different tests, different sampling, different ages, different age cohorts. And it remains to be seen if Australia “catches up” in the move from age nine to 15.
But none of this alters the primary observation that within the country, teachers have less to do with differences among students’ reading achievement than often supposed.
Shifting the debate
Australia’s former top public servant Terry Moran recently said the key to improving education standards is not in spreading more money on schools “like Vegemite”, but instead the focus should be on recruiting and investing in bright school graduates to become teachers and future school leaders.
Although it should be clear by now that we do not agree with any implication that teachers are the primary source of differences among students, no-one could object to the aspiration of attracting the most able people into the teaching profession.
For that to happen, teaching needs to be a respected calling. But a culture of blame and antagonistic press coverage undermine the respect that teachers need.