# Percentage of mistakes in stats;Letter

I WAS very disappointed to read in an article by John Howson a fallacious and statistically inept piece of writing (TES, September 9).

Mr Howson compares the 41.2 per cent rise in 15-year-olds achieving five or more A*-C grades, with percentage rises for other levels of achievement. If the rise achieving five or more A*-G grade were to match this, 112 per cent would have to reach the standard, and for those achieving 1 or more A*-G grades, 130.6 per cent.

To be statistically valid, Mr Howson needs to take a negative approach, examining instead the fall in the percentage of those failing to achieve each standard. Those failing to achieve five or more A*-C grades show a decrease of 20 per cent, those failing to achieve five or more A*-G grades show a decrease of 39.6 per cent, and those failing to achieve one or more A*-G grades show a decrease of 12 per cent.

In addition there is a good case to be put for assuming that a "perfect" score ought to be considered as 98 per cent, because it is often claimed that the percentage of children with severe learning difficulties is around 2 per cent. If this were taken into account, the percentages would become 20.7, 43.9 and 16.4 per cent respectively.

Reverend David A Prime

70, Beacons Park

Brecon, Powys

It only takes a moment and you'll get access to more news, plus courses, jobs and teaching resources tailored to you

## Latest stories

18 June 2018

Michael Tidd
18 June 2018

Helen Ward
18 June 2018
• ### £22m construction fund to pay for on-site training

Julia Belgutay
18 June 2018
• ### New UTC approved after DfE 'pause'

Julia Belgutay
18 June 2018

Kate Parker
18 June 2018

Helen Ward
18 June 2018

John Roberts
18 June 2018

FErret
18 June 2018
• ### Need to know: Grammar school expansions

Hélène Mulholland
18 June 2018