Dear Mr Gove

Michael Rosen is a perceptive man. He thinks and cares about state education deeply. Compare and contrast with the current incumbent at the DfE.

Dear Mr Gove: let's take a more scientific approach to looking at your education interventions

What evidence is there of a 'failed generation'? Precious little, but that doesn't matter to you, says Michael Rosen

This week I thought we could think about science; I'm not talking about science in the curriculum here, though what I'm talking about would be a good subject to study in schools. It's the science – or non-science – in your brain.

You take actions. At face value, you take these actions because you have made observations of the world and concluded that in the corner of the world where you have absolute power you could take actions that will change this world for the better.

Science relies a lot on observations, though traditionally has always found it difficult to distinguish between "what is observed", "how we observed it" and "how we described the observations". Some would argue that it's impossible to distinguish one from the other. You and your officers frequently collapse all three into one, expressed in one form or another as the idea that all state education in England, after the 1950s and before you came into office "failed". As if in proof of this assertion we hear from Sir Michael Wilshaw, talking of a "failed generation". It's an interesting idea, but one that would benefit from some science. This "fg", if it exists, would be aged roughly between 20 and 50. Straight question: is there any evidence to suggest that this age group is in any sense a "failure"? Is there any evidence to suggest that as a generation the cause of this "failure" is down to the education people received in the state system?Taking a "macro" view of things, the two major failures I see are the Iraq war and the Great Banking Crisis of the early 2000s. As far as I can discern, the people behind these two great generational failures didn't receive a state education. Do you and your officers have in mind some other great failure that this generation is showing us?

But let's stick with the idea that this is the "failed generation". The argument that you and your officers and approved thinktanks put out is that the cause of this "failure" is "child-centred education". Another straight question: how much child-centred education actually went on between, say, 1970 and 2010? A bit of science here: if there wasn't much, then hardly any outcomes – good or bad – can be said to have been caused by it. Both on personal observation of my own children, my school visits and in any accounts I've read, there was indeed very little of what is real child-centred education going on in this period.

Now we come to the action bit. This is what some scientists call an "intervention". Having made the observations and having figured out the cause for the state of affairs, you intervene. Now, science would argue for a bit of caution here: first a pilot; then an evaluation; then, if all good, the rollout; if not, abort.

So (not in chronological order):

Intervention 1, from the Gove lab: Free schools. Pilot? Nope. Evaluation of pilot? Nope.

Intervention 2, from the Gove lab: Rollout of free schools.

Intervention 3, from Labour: Academies. Pilot? Yes. Evaluation of pilot? Yes. Any problems with the evaluation? Yes. What was evaluated were individual schools, each of which were special cases. If we say the objective is "raising standards", that holds within the phrase a sense of "raising standards for all". So, even if the academies in question raised standards for the students at those academies, we can only say "raises standards" if we look at the whole cohort in any locality, ie everyone. Was this done? Nope.

So, intervention 4, from the Gove lab: Rollout of academies. Now the Gove evaluation: As expressed in your last major speech, which said that all this has been a great success.

But there's a problem. This week we have heard that up to 10 academies are are having to be relinquished by one of the academy "chains", and you are looking for new sponsors.

At this point, a scientist would have some urgent questions:

1 Is the failure of one-third of a chain's academies statistically significant? If so, would it lead us to think that the failure is not with the academies as such, but with the chain that was running them?

2 If a whole series of schools could fail like that, is it a problem with that chain alone, or is there something wrong with the models of a) chains and/or b) academies?

Now, correct me if I'm wrong, but I'm under the impression that you think that everything I've written here is irrelevant tosh. That's because you don't believe in science. You believe in a) your opinion and b) your power. You slot your opinions into the enactment of power and, hey presto, we have “policy". (my emphasis)

In these circumstances, who needs science? And who needs democracy? Not us, it seems.

Yours, Michael Rosen

Guardian 4/3/14