Next month, Nesta is bringing together leading experts, policymakers, data scientists, civic society groups and frontline practitioners from across Wales and the wider UK for the latest ‘Using data in the early years’ conference.

In anticipation, we thought it would be useful to consider the reflections of Professor Leon Feinstein, the keynote speaker from our last conference, whose insightful analysis foregrounds many of the discussions we’re hoping to have this time.

Over 45 minutes he covered a wide range of topics from the conceptual, such as the politicisation of evidence, to the specific, such as the excellent local work in Oldham and Rochdale. It was a talk that reflected the experiences of many of the attendees, especially the many challenges faced when trying to wrangle with messy, imperfect and siloed datasets to improve children’s lives. As he set out the bigger picture, there was a collective realisation that these were not unique, individual challenges but ones shared throughout the audience, across all of the early years.

Personally, a number of his points resonated, reflecting Nesta’s work in this field over the last few years. For example, we repeatedly see the challenges faced by councils in sharing data and using data better and the nodding heads showed this was not just the experience of the councils we’ve worked, but also the many local authorities, charities and not-for-profit organisations that attended our event.

Most importantly we need the right data.

Quoting Peter Riddell, the former director of the Institute for Government, Prof Feinstein observed that “when the government changes, the evidence changes”.

Data needs to reflect practice rather than what Whitehall wants. Through our many conversations with early years services, we’ve come to understand how many datasets are bloated with data being recorded for a previous government's requirements.

That’s why we think a “spring clean” of data should be done regularly to ensure the data being collected helps make local decisions at a local level, rather than reflecting what was once considered “evidence” by a past government.

The data the centre needs isn’t the data local needs

Prof Feinstein gave the example of the S251 dataset - the annual budget and outturn for services for children, young people and education - a dataset that contains huge errors precisely because it has no usable value to councils.

At Nesta, we’ve seen this similarly within child development data too. For example, the Ages and Stages Questionnaire (ASQ-3) is a population-level measure, useful for the Department of Health and Social Care to track outcomes at a national level, but not so useful for local authorities (LAs) trying to put the correct support in place based on a child’s individual needs and circumstances. But councils end up using ASQ-3 to screen for developmental issues at an individual level, something which the tool is not designed to do, because this is the data they have to collect for central government, even though it’s of little use to them.

As a result, we are increasingly seeing them use other measures alongside the ASQ-3; we found from our Mapping Early Years Practice pilot that 14 different assessment measures were being used in nine different ways across 27 LAs. Considering there are 152 LAs in England alone, we can only assume that there are even more assessment tools being used, precisely because the ASQ-3 is useful for Whitehall, but not for LAs themselves.

Government departments don’t align with children’s lives

Children and families exist in their own context, rather than through the lens of the DHSC, or the education and work and pensions departments.

And that disparity is a root cause of the data problem. Each department requires different data for their own evidence, which then changes when a new secretary of state comes in. From their perspective, the families are siloed and don’t exist across departments. During our work in early years we encounter this separation of services in the early years time and time again, across the UK.

It’ll take time to make changes to the use of data in the early years.

Prof Feinstein mentioned not being drawn in by “quick wins”: producing a dashboard without useful insights won’t help in the long run.

But if we invest the time, it could also save time, provided it’s the right data.

He explained the need for data to link to practical decisions and actions, so that local authorities have the information they need to intervene properly.

Currently we have a paradox. We have a prevalence of wrong data, with services having troves of data that has no utility to them, but neither the time or capacity to collect the extra data that could actually be helpful - despite there being a desperate need for better data to inform decisions, understand the critical wider context and make sure children and families are getting the support they need.

It’s undoubtedly true that if we could intervene earlier, we could save time, resources and money later on.

There’s clearly a disconnect between the centralised government department data view and the real world, with Prof Feinstein highlighting that the “distance between what’s demanded and what is possible has never been wider”.

For me, this reminded me that collecting data is ultimately a manual process. The more data that’s asked for, the more time a front-line practitioner has to spend collecting and inputting it.

Part of a health visitor’s job is asking the questions to families to get the data and then inputting that into their IT system. When the data is sent across to the government departments, or indeed to Nesta, in large spreadsheets to be analysed, it is too easy to forget that someone had to spend time creating that data.

At a time when health visitors and the early years workforce are struggling with increasingly unmanageable caseloads combined with increasing staff to children ratios as childcare entitlement expands, we need to remember, that any extra time doing admin, is less time seeing the children and families, and more pressure and stress on overstretched staff.

Much of what Prof Feinstein says we’re advocating for at Nesta - the importance of data accuracy, relevance, and the inclusion of voices from practitioners and families and joined up data across services. It was a call for action - a holistic approach to data collection and analysis is needed. We need to prioritise quality over quantity and to empower those on the ground so we can have effective, impactful practices moving forward.

Nesta will be hosting the next data in early years event on Monday 18 November in Cardiff.

If you're interested in finding out more and joining us for this session, please email [email protected] with the subject line "November early years data conference".