Data lessons from the education sector

When comparing the vast number of government buildings in London, a converted church in High Holborn is probably one of the nicest. It is at this address that Ofsted (the education regulator) has its offices. Much nicer premises than the slightly ugly 1960s tower the Care Quality Commission calls home. 

‘The Department for Education was much more open to our ideas’

It was at the Ofsted offices that two health information professionals found themselves in August of this year discussing, of all things, a good GCSE guide they had been commissioned to produce. “Why a healthcare team?” was the first question from the Ofsted board member? A good challenge.

About a year ago, the Open Public Services Network asked Roger Taylor, a director at Dr Foster, and myself to review the data currently being published by the Department for Education and see if more could be done to analyse and interrogate this to make it more accessible for parents, pupils, school governors and government. It was struck by work we had done previously for Dr Foster and NHS Choices, presenting complex information to patients and the public.

Working with specialist analytical companies RM Education and FFT (and with funding from Cambridge Assessment), we produced a report explaining how education data can and should be used to choose schools. This involved applying techniques, such as banding of outcomes, which is done routinely in health when measuring mortality rates. This was supported by The Guardian, which published an online “GCSE schools guide” to help parents find and research local schools.

Cultural differences

But while the application of technical aspects across the sectors maybe interesting, it is the cultural differences between the two public services that are worthy of discussion and comparison.

I have been working with health organisations publishing data for about 10 years, both on behalf of non-governmental bodies and private sector organisations, and these have almost always been met with suspicion by the Department of Health. A “not invented here” attitude, a reticence to hand over the data and a fear of the consequences of publication.

‘The health sector does not make access to information easy’

The Department for Education on the other hand was much more open to our ideas, with senior officials meeting with our expert panel (which oversaw the research) to discuss and debate our approaches. The data we needed, the national pupil database, was supplied in an easy and timely way under a licence with reasonable requests (such as that we previewed the analysis to the department and coordinated launch activities). 

This, I think, is contrasted by the health sector, which does not make access to information easy. We cannot, for example, have proper access to national joint registry or clinical audit information and the great transparency success of the summer − the publication of consultant level outcomes − tended to be in a non-machine readable PDF format meaning researchers were prevented from creating comparable projects to the GCSE guide.

Bastion of transparency

The Health and Social Care Information Centre is not doing enough to enable credible organisation access to linked deaths, which will make measuring mortality more accurate. Even the cardiac surgeon’s website − often heralded as the bastion of health transparency − lay dormant for three years as the various agencies argued over who would fund it.

And it is these “various agencies” in health that also cause problems. In education there is the department and the regulator, but in health there are numerous bodies − the DH, the information centre, NHS England, Monitor, the CQC, the NHS Trust Development Centre and so on, which will take an interest in a publication, not be willing to coordinate with counterparts and in some case be very obstructive to the publication. 

‘Both sectors have organisations that leave data gaps but at least the DH mandates data collection from foundation trusts’

Of course it is not all one-way traffic, and you may point out most of my health experience was before the changes on 1 April. NHS England does have a national director, Tim Kelsey, with a remit and a passion for transparency. There are also some types of data collected in the NHS that the education sector could benefit from. This includes teacher and pupil surveys, taking a lead from the NHS staff and patient surveys, and a variant of the friends and family test.

Both sectors have organisations that leave data gaps − private hospitals in the healthcare sector − but at least the DH mandates data collection from foundation trusts. This is not always true of for academies and free schools − private schools supply very little information to the Department for Education. 

There are many more pieces of information the education department could release, and searching for “health” returns more data than “education” on data.gov.uk. I also understand it is healthcare sector’s pseudonymisation techniques that allowed the national pupil database to be made available in the first place. 

So in answer to the question from Ofsted, why a healthcare team? Perhaps it was the battle wounds from navigating health that made us all the more ready to take on education.

Alex Kafetz is a director at ZPB and a member of the Cabinet Office open data user group