Department of Education

Viewing archives for Data

Alastair Lee

This blog post is written by Alastair Lee, Children’s Services Data and Information Manager, East Sussex County Council  and Chair of the Children’s Services National Performance and Information Management Group. It is part of a series published by the Rees Centre on data.

Data Tools in Local Authority Children’s Services

In every local authority there are people using data to monitor the performance of services. Ideally this supports a conversation between managers and frontline staff to help them get a better understanding of the demand on the service, the pressure on staff and the impact on children, young people and families being supported.

Producing the data in an easily understandable format, whether this is as tables, charts or dashboards does not happen at the press of a button. A lot of data wrangling is needed to get the data in a fit state to feed the reports.

Data wrangling

Most of the data we get doesn’t come in a useable format. As a result we need to manipulate it to combine multiple reports, add a specific data column or extract specific data to meet our needs. The most basic approach to this is cutting and pasting, but as this can become very long winded and tedious we have developed tools that speed this up. Whether using advanced Excel formulae, VBA or SQL or the specific data manipulation aspects of Tableau, PowerBI, Knime, Alteryx or R the basic aim is the same. Build something that speeds up the process, reduces the tedium and as a result improves accuracy. My team have a shared folder where these tools are kept and whether used weekly, monthly or annually they are incredibly valuable.

Presenting data

Once we have the data we need in the right format to use we then plug it into a data visualisation tool. Currently the programme most often used for this is Excel simply because all staff have it on their PC or laptop and so it is easy to deploy widely. Also if an Excel dashboard is developed for a specific purpose (e.g. the ChAT[1] or the Children’s Social Care Benchmarking tool that were developed by the Data to Intelligence project) I can download it, add my own data and get an output without any need to install any new software. But Excel has limitations when working with large, complex datasets and in how the data can be visualised. As a result work is being developed using more bespoke programs (e.g. Tableau, QlikView and PowerBI) and once a local authority has these installed and a way of deploying to colleagues, the same ability to share templates and visualisations is true.

Analysing data

When analysing data Excel is still the most common tool used, it’s what we’ve got! Some local authorities use SPSS and R is beginning to appear. This is being driven by expertise arriving with new staff who have used these programmes elsewhere. The analysis is being driven by a greater interest from service managers about longer term impact of interventions/services, the impact of changes to services and the need to forecast more accurately.

A place to share

The situation at the moment is that many local authorities have developed their own data wrangling, visualisation and analysis tools and some won’t because they don’t have the people with the skills to do so. This leads to duplication of effort between some local authorities and other not accessing tools that would help improve outcomes for children and young people. This is a waste!

To address this the Children’s Services National Performance and Information Management Group (CS-NPIMG), the South East Sector Led Improvement Programme (SESLIP), the Data to Intelligence project, Ofsted and Social Finance’s Collaborative Technology Initiative are developing a curated data tools library where these tools can be hosted, shared and co-developed. The project is in its very early days but we have good learning from the open source movement, and from the development and sharing of the ChAT which is now used by 150 LAs and the SE Data Tools library which has looked at the impact sharing a tool can have on the local authority that shares it.

One unexpected consequence of our current experience of sharing tools is that it improves data quality in the statutory returns.

This happened because it enabled us to see what the data would look like once processed by the Department for Education, previously we’d only know this once a submission was made and an error report was returned. As the tools are open source, we can all see how the data we enter is being transformed to create the output and this can reveal where errors may have arisen in the past. This has also contributed discussions about the development of standard data sets that we can all use for analysis, visualisation and research. This is all from a very limited number of shared tools; there is more work to be done to increase the number of tools for visualisation, data wrangling and analysis that will help improve outcomes for children and families in need of support.

[1] The ChAT is the Children’s Services Analysis Tool that was developed by a group of London LAs and Ofsted to better visualize the data that is shared between and local authority Children’s Services department and Ofsted during an inspection.

This blog post is written by Alastair Lee, Children’s Services Data and Information Manager, East Sussex County Council  and Chair of the Children’s Services National Performance and Information Management Group.
Contact Alastair: Alastair.Lee@eastsussex.gov.uk

It is part of a series published by the Rees Centre on data.

Related network:

Children’s Social Care Data User Group

The Children’s Social Care Data User Group (CSC DUG) was set up in 2017 by the Rees Centre, University of Oxford and Thomas Coram Research Unit, UCL. It is a network of academics, local authority data managers, analysts, charities and funders with a shared vision that administrative data from children’s social care and other relevant agencies in England can be analysed and fed back into policy and practice to improve the way that children’s social care services respond to children, young people and their families.

The group focuses on data submitted to and analysed by the Department for Education (namely the SSDA 903 children looked after data and the Children in Need Census data).

Membership is open to any individual or organisation who is using, or plans to use, children’s social care data collated by the Department for Education (Child in Need data, Looked After Children data or Section 251 social care expenditure data in relation to children’s services).

To join the group’s mailing list: email rees.centre@education.ox.ac.uk

This insight piece by Dr Lisa Holmes for the Nuffield Family Justice Observatory explores the nature, availability and use of child level administrative data at the local and regional level by children’s social care departments.

“It is evident that there is value in children’s social care data being analysed locally and regionally, as well as purely being exported by local authorities to meet the statutory reporting requirements, but that the capacity and capability of performance management teams impacts on the ease and frequency that this can take place. “

Use of these data are considered within the context of requirements to submit annual administrative data to the Department for Education. Consideration is also given to issues related to the capacity of local authority performance management teams which have been drastically reduced in many local authorities as a result of austerity.

Finally consideration is given to ways that local authorities may share learning and increase capacity and capability.

Full report Nuffield Family Justice Observatory 21/10/2019

 

This year’s annual International Association for Educational Assessment (IAEA) conference, hosted by the department’s Centre for Educational Assessment, welcomed over 400 participants from examination boards across the globe to St Catherine’s College, Oxford from 9 – 14 September.

The theme for the conference was Assessment and Big Data, with keynotes delivered by Rebecca Eynon (Department of Education), Art Graesser (University of Memphis) and Michelle Meadows (Ofqual), which focused on the social implications of Big Data and data tracking and the use of intelligent agents in tutoring and assessment, to the opportunities and limitations of data as a regulatory tool.

The conference also included an opening address from the University’s Vice Chancellor and an introduction to the conference theme by the department’s Director, Professor Jo-Anne Baird.

Watch all the conference highlights here.

For more information about the department’s Centre for Educational Assessment see: http://oucea.education.ox.ac.uk/