Tag: LMI

Getting Universal Jobmatch stats

Archive post: Universal Jobmatch was replaced by Find a job on 14 May 2018.

Have you tried to use Universal Jobmatch to access data on job vacancies? If so, you may have been shocked and dismayed by what you found.UJ logo2

Universal Jobmatch (UJ) is intended to be a “new best of breed online service” for jobseekers and employers, intended to transform DWP’s labour market services, automatically matching jobseekers to jobs based on their skills and CV. It is run by Monster, on a £15m contract over four years. It is a potentially very valuable source of data, and the contract specification included requirements for LMI summary tools. UJ data replaced those from the Jobcentre Plus notification system which ceased in November 2012, and was expected by the Department of Work and Pensions to provide a more comprehensive picture and offer improvements in how the data can be accessed and used. Possible UJ analyses include vacancies by industry and occupation; numbers of employers with vacancies; and qualification levels amongst jobseekers and levels required by employers.
UJ reports home page

However, closer examination of UJ raises many concerns, several of which have recently surfaced on the Labour Market Statistics Group on StatsUserNet. The serious deficiencies are very unfortunate at a time when there are increasing concerns about skills mismatches, the quality of jobs and careers advice, and LEPs working with partners on skills strategies and local EU growth programmes.

First, there are issues about how the UJ statistical reports function works – with no guidance and descriptions of the data (metadata) provided. Drop down menus for local authority areas stop part way through the alphabet (tough if you’re after East Devon on one and Kingston upon Thames on another). You can derive top ten rankings, eg, for occupational groups – which generates a bar chart where you need to hover your cursor to see what the bars contain. You have the option to download into Excel, and all this provides is a picture and no data.

monthly trend reportSecondly, UJ does not use conventional classifications, for geography, industry or occupation. The contract spec for UJ expected the use of Standard Industrial and Standard Occupational Classification (SIC & SOC) codes, but UJ currently uses Monster’s own taxonomy. Monster’s US origins are evident in the report on qualifications held by jobseekers where ‘Some High School Coursework’ and ‘High School or equivalent’ are categories. User interfaces are, however, tailored to Britain: when employers post vacancies, they are asked to select job location regions, though these correspond more to ITV broadcasting (Anglia, Tyne-Tees, etc) rather than administrative regions. Analysis of the industry breakdown of vacancies is made problematic by the high proportion which feature under ‘Staffing/Employment Agencies’, rather than the sector concerned. Looking at data for Hertfordshire in March, for instance, showed 55% of all vacancies in this category.

Thirdly, UJ is typically capturing a different profile of vacancies than the previous JCP system, with many more managerial and professional posts and many fewer low skill or no skill vacancies.  There are also oddities which seem difficult to explain: the StatsUserNet discussion, for instance, commented on numbers of vacancies for ‘diplomats’ in the Tees Valley and Cumbria.

It’s important to understand how the new system works, combining as it does posts entered by employers, bulk uploads from ‘job warehouse’ sites, and other vacancies ‘scraped’ (by agreement) from the Internet. In this, there is relevant learning to be gained from the USA, where the Department of Labor has supported projects which made use of data using online job postings for real-time LMI. This experience offers a good number of warnings, for instance about the likelihood of duplicated postings, under-representation of jobs in smaller companies and in rural areas, and of variations by sector, locality and level of geographic analysis.

It’s sad that DWP and Monster have not adopted a more open and inviting approach to improving the site – it’s far from co-design principles that get the best out of interactive websites. In the August 2012 DWP Quarterly Statistical Summary, there was an article which explained DWP’s new approach to vacancy statistics, inviting comments and referring to a six-month beta stage.

There was a statement in Parliament (Hansard, 4 June 2013) that DWP and Monster had a timetable for prioritising and implementing improvements, but this had not been made public.

Finally, while these are gripes from a labour market analysis perspective, spare a thought for jobseekers. It’s not difficult to find criticisms of how the system works for them. It’s very serious: Jobseekers Allowance claimants could be required to look for work using Universal Jobmatch , or risk losing their benefit.

Permanent link to this article: https://www.educe.co.uk/?p=1183

Doncaster Work & Skills Plan

Derrick Johnstone advised Doncaster Council and the Work and Skills Steering Group of Doncaster Together on their Work and Skills Plan. This built on the Local Economic Assessment and set out to make the most of the added value of partnership working, in the context of far-reaching changes in funding and national policy. It provided the basis for discussions on the introduction locally of the DWP Work Programme, and prioritised action around employer engagement; labour market intelligence; information, advice and guidance; and targeted support for individuals and families. It has also informed the work of South Yorkshire Local Enterprise Partnership on skills.

Permanent link to this article: https://www.educe.co.uk/?p=278

Reducing Worklessness in Norfolk

We assisted Norfolk County Council and partners in developing the Worklessness Assessment and Framework for the county, in partnership with OCSI and Papworth Consulting.Elements of the work included:

  • analysis of changes in the Norfolk labour market, needs and barriers amongst disadvantaged groups
  • customer research, focusing on the needs and experiences of individuals from different groups and parts of the county
  • pulling together information on current worklessness provision in the county, building on a regional mapping project undertaken by the University of Glasgow
  • reviewing the implications of a changing policy environment
The Worklessness Assessment fed into the Norfolk Economic Assessment and the County’s Child Poverty Assessment and Strategy. The work subsequently provided a basis for influencing the delivery of the Work Programme in the county.

Permanent link to this article: https://www.educe.co.uk/?p=258

Data sharing on worklessness

One of the key issues affecting the efforts of local partners to tackle worklessness relates to constraints on data sharing. Following the Tackling Worklessness (Houghton) Review, CLG and DWP agreed to initiate a pilot project to demonstrate ways in which relevant data can be shared within the current legislative framework. The pilot involved Kent, Leeds and Liverpool City Region, and aimed to clarify barriers, enabling factors and ways forward.Derrick Johnstone led input by three Local Improvement Advisors (LIAs) in support of the pilot areas, also contributing as a member of an Expert Group. The role of the LIAs was to:

  • broker relationships and facilitate discussions around local needs for worklessness data
  • facilitate data sharing and data management to achieve better outcomes
  • help identify solutions to obstacles, including ways of ensuring data security
  • working with local partners, CLG and DWP to test and provide exchange protocols, data sharing principles, Memoranda of Understanding, and other tools to underpin this process
  • identify lessons and good practice to be shared more widely as outcomes from the project.

The project has informed guidance on data sharing and worklessness published by DWP, and the IDeA ‘How To’ guide on Good Practice in Data Sharing ( 840KB) drafted by Educe. It has also led to DWP providing working age benefits data at very small area level (Output Areas), the smallest areas for which statistical data can be supplied.

CLG have published an evaluation of the pilots, Sharing data to improve local employment outcomes ( download on CLG site).New possibilities for data sharing have since been opened up by the Sections 130-133 of the Welfare Reform Act 2012. Draft regulations have been published to extend data sharing to help identify people affected by new benefit rules, especially on housing, and families with multiple disadvantages who may require support to help turn their lives around, as part of the Government’s Troubled Families programme.

Permanent link to this article: https://www.educe.co.uk/?p=241


Educe worked with OCSI (Oxford Consultants for Social Inclusion) on the further development of the  well-regarded Data4nr (‘Data for Neighbourhoods and Regeneration’) website, under commission to the Department for Communities and Local Government. Data4nr identified and signposted the datasets available for targeting, monitoring, priority setting and performance management at a neighbourhood level, also highlighting – where possible – sources which provide equalities data. Our role has been to advise and support on user engagement in developing the site, and help feed in new material.

Data4nr was used to kick start content in the development of the Government’s Open Data site.

Permanent link to this article: https://www.educe.co.uk/?p=220

Reducing Inequality Review

The Reducing Inequality Review, for Brighton & Hove Council and the 2020 Community Partnership (LSP) analysed existing evidence on dimensions of inequality in the city, across neighbourhoods, groups and communities of interest, before proceeding to review the contribution of local policies in reducing inequalities and disadvantage and consider future priorities. The review was undertaken jointly by Educe and Oxford Consultants for Social Inclusion (OCSI), and carried out in two phases:

  • Phase one was essentially a needs assessment for the city, intended to set the scene for Phase two.
  • Phase two offered recommendations for approaches and strategies for reducing inequality in the city, which fed directly into the preparation of the new Local Area Agreement (LAA)

We worked with a steering group made up of various stakeholders, with the project involving over 100 people in the research, soundings and evidence gathering. The Reducing Inequality Review brought together for partners, for the first time, a substantial range of evidence to enable a coherent and up-to-date understanding of the needs of the city. The Council regard the review as an example of best practice in how cities can develop a sense of their local identity and the opportunities available.

The Phase 1 and Phase 2 reports are available for download from the Brighton & Hove Council website.

Permanent link to this article: https://www.educe.co.uk/?p=163