Levels of response marking

Some answers cannot sensibly be marked by ticking. An essay comparing two poems is not better because it contains more facts; it is better because the argument is more perceptive, the evidence better chosen, the analysis more precise. Levels of response marking exists for exactly these answers: it describes what each standard of work looks like, and asks the marker to judge which description fits.

Best fit, not a checklist

The governing principle is "best fit". An answer rarely matches every phrase in a band descriptor - it may be strong on analysis and thin on context - and the marker's job is to decide which band it fits overall, not to demand every feature before awarding it. Once the band is chosen, the marker places the answer at the top, middle or bottom of that band. Treating a levels-based descriptor as a checklist is the most common way this method is misapplied, and it systematically depresses marks for uneven but strong work.

Why it is harder to apply consistently

Because the judgement is holistic, two markers can place the same answer in adjacent bands and both be defensible. That variation is not a failure of the method; it is the cost of assessing something a checklist cannot capture. It is also why departments marking extended work standardise beforehand against exemplars and moderate afterwards - the descriptors alone are not enough to guarantee a shared standard.