Gerard Kelly, The TES editor, defended “social engineering” in his editorial last week by writing that “despite the impressive performance of independent schools at A-level, state pupils maintain a slight edge when it comes to the quality of final degrees”.

In one sense he is right - state school pupils with the same grades at A- level achieve better degree classmarks on average.

But in another sense he is wrong - the average independent school pupil who goes to university gets a better classmark than the average state school pupil, the same research studies show. So this implies that admissions tutors are indeed engaging in social engineering in favour of state school pupils - but they may be doing it too much, if we measure their success or failure by judging how good they are at selecting pupils who will earn good degrees.

This fact is glossed over by education experts who conduct the research. This reflects a broader problem - that much education research is so heavily politicised as to be worthless. I’m all for social engineering, and Mr Kelly’s editorial piece makes some intelligent points on this. But it has to be justified by the results - in this case the result that the right talent has been spotted.

• David Baresch, Wandsworth, London