EXCLUSIVE: UK Stats Watchdog again expresses concern about a DfE announcement
It’s happened before. And it’s likely to happen again. The Department for Education (DfE) publishes an announcement with misleading data. This is widely churned by the media.
By the time complaints are investigated by the UK Statistics Authority (UKSA) the world has moved on. And the misleading data continues to be accepted as the truth.
Take the DfE press release about results in the five-yearly international reading test for 10 year-olds (PIRLS). English pupils did well – up from joint 10th in 2011 to joint 8th in 2016.
This was hailed by the DfE has ‘a dramatic improvement on the 2006 results, when England was ranked 19th out of 45 countries’.
But the ‘dramatic improvement’ had been between 2006 and 2011. The rise in ranking from 2011 was an improvement, yes, but not as great as the jump from 2006.
Leaving out the 2011 results was a statistical sleight of hand as I wrote here. Their omission allowed schools minister Nick Gibb to claim the improved results were due to ‘an increased emphasis on phonics’.
I complained to UKSA. The watchdog has now replied. It said they were ‘satisfied’ the DfE had taken ‘appropriate steps to ensure the statistics used were consistent’ with the PIRLS National Report for England published jointly between the Oxford University Centre for Educational Assessment (OUCEA) and the DfE.
‘…the news release only references England’s 2006 and 2016 PIRLS results. In our view, a more complete picture about ongoing improvements in education and the impact of specific policies would have been provided if England’s 2011 PIRLS figures had also been included in the release.’
UKSA has written to DfE’s Chief Statistician to highlight these concerns.
It’s too late. TES, commenting on Nick Gibb remaining at the DfE, wrote:
Mr Gibb's stock is high in Conservative circles after last month's Progress in International Reading Literacy Study (Pirls) rankings put England joint-8th in the world for 10-year-olds’ reading…’
But ‘Conservative circles’ should take note:
- UKSA has concerns about Gibb’s presentation of data. And not for the first time.
- The joint DfE/OUCEA report said it was too hasty to say policy changes since 2010 had contributed to England’s rise in PIRLS.
- England’s PIRLS score in 2016 was ‘significantly below’ Northern Ireland. There does not seem to be the same emphasis on synthetic phonics there: see here and here (scroll down). Northern Ireland puts its emphasis where it should be – on teacher professionalism.
NOTE: UKSA correspondence can be downloaded here. Look for letters dated 11 and 12 January 2018 re PIRLS results and my original complaint on 5 December 2017.