Experimental trials and ‘what works?’ in education: The case of grammar for writing

Methodological re-examination of two influential experimental trials plus a review of the wider grammar-for-writing evidence base

English national curriculum policy for writing, ages 5-16

Methodological critique published in the British Educational Research Journal 43(6), examining in detail two well-regarded and policy-influential trials of grammar teaching for writing, and setting them against the rest of the field and a then-recent robust RCT. The value of this source for the archive is that it is a specific, named adversarial pass on the single positive finding in the topic rather than a general commentary.

The critique identifies concrete design defects in the trials that drove policy - notably that intention-to-treat analysis was not used and that analysis was conducted at pupil level rather than at the level at which randomisation and delivery occurred. Its broader conclusion is a documented mismatch between national curriculum policy in England (which expanded grammar requirements sharply from 2014 and made them statutorily testable through the KS2 GPS test) and the robust evidence available on the teaching of writing. The paper is the source of the caveat that travels with Jones/Myhill's e = 0.21 wherever it is cited.

Not applicable - a methodological critique with no estimates of its own.