Good teaching matters. We know this rhetorically. McKinsey’s Report How the world’s best performing schools come out on top asserted: ‘The quality of an education system cannot exceed the quality of its teachers.’ The Cambridge Primary Review affirmed that, ‘teaching makes a difference. Excellent teaching can transform lives.’ Gove’s inaugural White Paper was entitled ‘The importance of teaching’.

It’s not a new discovery. The 1959 Crowther Report asserted that ‘Everything in education depends ultimately on the teacher.’ The historian Tony Judt observed that ‘being well-taught is the only thing worth remembering from school.’ President Obama claimed that ‘the most important single factor in determining achievement is … who their teachers are.’

We also know this empirically: some impressive mathematics have been deployed to isolate the teacher effect. Rand Education concluded that teachers matter more to student achievement than any other aspect of schooling. Teachers have two-three times the impact of any other school factor – including services, facilities and even leadership.

The Sutton Trust calculated that during one year with a very effective maths teacher, pupils gain 40 per cent in their learning. Another study found that being taught by a high quality teacher adds 0.425 of a GCSE point per subject.

Factors that contribute to good teaching

It’s one thing to establish that teaching is crucial to student achievement, but quite another to recruit the explanatory algorithms for the purpose of evaluating teachers. The econometric approach ignores a crucial set of factors that contribute to the success of teaching, but over which teachers have little control.

Attempts to isolate the impact of teachers start with student baseline scores, and then try to apportion progress between a set of factors that includes the ‘pupil effect’, the ‘teacher effect’ and the ‘school effect’. But what exactly is the ‘school effect’? In some studies, this includes resources and facilities, which might be assumed to vary between schools. Other studies control for ‘classroom effects’ – covering class size, peer characteristics and ability range.What can’t be explained then gets reduced to a ‘residual’ in regression equations.

These studies do not satisfactorily isolate the set of conditions within which teachers work, and which contribute to success. Putting aside resources and school leadership, teachers have to work with given timetable allocations, length of lessons and transfer time, independent study conditions, quality of a school’s virtual learning environment, nature of cover arrangements, school assessment and progress monitoring policies, the compulsory or optional status of a subject, the changing nature of exams, and the quality of middle leadership.

How do you compare teachers who have a fixed classroom base with those who are constantly having to move between rooms? Being able to configure a classroom to suit a particular purpose is by no means a given.

Many of these structural and logistical factors affect teachers differently even within the same school. Many will vary (if not by policy, certainly by consistency of implementation) by subject and stage within a school. Teaching matters, to be sure, and so do teachers. But a teacher’s impact evidently depends on so much that is outside of his or her immediate control.