New Early Years stats publication – joint working with NHS Digital
Telling the story in early years
As part of Ofsted’s Early Years Data and Analysis team, I help produce official statistics about the number of early years and childcare providers in the sector and their inspection judgements.
In 2017 we set ourselves the challenge of investigating what other datasets about 0-5 year olds were available across the GSS. We knew about the Department for Education data on outcomes at 5 years (EYFSP) but soon discovered the 2-2.5 year health review – one element of which is the Ages and Stages Questionnaire (ASQ-3).
The Ages and Stages Questionnaire
Those of you with young children may have received a letter inviting you to complete the questionnaire with a health visitor. It assesses whether a child’s development is on schedule in five domains: Communication, Fine Motor, Gross Motor, Personal/Social and Problem-Solving. If a child falls ‘below threshold’ in any of these domains, their health visitor may make a referral to a hearing specialist or speech and language therapist, for example.
We soon became excited at the prospect of linking up ASQ-3 data on 2-2.5 year olds to Ofsted data on early years settings and potentially even the DfE’s EYFSP data. In theory, if we could find suitable matching variables, we could investigate whether children who fell below threshold at 2 also failed to reach a ‘good level of development’ in reception year – or if they didn’t fail, did they attend Ofsted-registered early years provision in the three years in between? Was it Outstanding or Inadequate provision? Having our hands on this data could help Ofsted measure the value of our early years inspection judgements. Our policy colleagues were also keen to support our work to understand and link GSS data.
Secondment to NHS Digital
After six months of meetings and sending briefings and reports back and forth, NHS Digital offered to host me on a one-month, part-time ‘honorary contract’ in their Leeds office. This operated similarly to the secondment scheme run by the GSS Good Practice Team, in that I effectively became an NHS Digital employee but was still paid by Ofsted. With help from colleagues, I broke down published data by demographics such as gender, ethnicity, first language and deprivation decile. We then compared ASQ-3 and EYFSP outcomes in 27 local authorities to see if there was a correlation between development outcomes of 2 year olds and 5 year olds in the same geographical area.
What I learnt
The biggest difference between working as an analyst at Ofsted and NHS Digital was how they handle the data. At Ofsted we redact the personal information of certain provider types prior to publication, but we don’t hold any child-level data. At NHS Digital, we needed to round every ‘number of submissions’ to the nearest 5 and suppress numbers below 5 so that individual children couldn’t be identified through the dataset. As a result, I’ve learnt some new Excel and SQL skills – I can now create temporary tables, joins and pivot tables.
On 12 September Ofsted and NHS Digital published a joint report into the 24, 27 and 30 month Ages and Stages Questionnaires. NHS Digital will continue to publish a breakdown of the data by child demographics so we can identify trends and changes over time.
Since October 2015 it has been mandatory for NHS-funded community services to send their ASQ-3 records to NHS Digital, but the data remains patchy. As interest in the dataset increases, we hope that more community health providers will adopt the systems needed to submit this data to NHS Digital. Once the data coverage is sufficiently strong, we can revisit our plans for a longitudinal study.
Ofsted and NHS Digital’s joint report into the Ages and Stages Questionnaire is available on gov.uk here.