Developers Guide

We are aware that many organisations are contracting developers to build apps based on LMI for All. Despite robust documentation of the API and detailed explanations of the data sources, it can be difficult to make sense of the significance and meaning of the different data sources and end points and how they can be used. In this (hopefully plain language) guide, we try to explain what you may be able to show with the different data and provide links to further possible resources.

Different users

There are many different potential user groups for data from the LMI for All API. Many of the existing Apps are targeting young people seeking careers information. Typically they might want to know about qualifications needed for a particular job, qualification routes and what skills and knowledge are required, how much pay they might expect to earn, and future employment and pay prospects. But different age users may also have different needs and expectations. Younger users may be trying to find out about different jobs, older users might be more interested in qualification routes and future opportunities. Different age groups may not only have different data requirements but may respond better to different ways of navigating and visualising data.

Of course the choice of design will also depend on how different organisations intend to use the LMI for All data. There is a big difference between an application intended to be used by an experienced careers guidance professional and one intended for free use by young people and end users. And whilst some organisations will want to use LMI for All as a stand alone application, others will wish to integrate it within their existing online provision.

The use of LMI for All is not just for young people. The data may be of interest to older unemployed people or to those wishing to change occupations. It could also be useful for local government planning departments or for those responsible for planning and developing college and university provision. In each case the end user may affect both the data selected for display and the design of the applications.

Occupations

Much of the data available through the LMI for All database is classified according to the UK Standard Occupational Classification (SOC). This system, last updated in 2010 by the Office of National Statistics (ONS). Within the context of the classification, jobs are classified in terms of their skill level and skill content.

It is used for career information to labour market entrants, job matching by employment agencies and the development of government labour market policies.

SOC2010 has a hierarchical organisation with nine major groups, 25 sub-major groups, 90 minor groups and 369 unit groups. Each group has a code with major groups having a one figure code, sub major groups a two figure code, minor groups a three figure code and unit groups a four figure code.

The following is an example of how the classification is structured:

Major Group Sub-Major Group Minor Group Unit Group Group Title
(1 digit) (2 digit) (3 digit) (4 digit)
1 MANAGERS, DIRECTORS AND SENIOR OFFICIALS
11 CORPORATE MANAGERS AND DIRECTORS
111 Chief Executives and Senior Officials
1115 Chief executives and senior officials
1116 Elected officers and representatives
112 Production Managers and Directors
1121 Production managers and directors in manufacturing
1122 Production managers and directors in construction
1123 Production managers and directors in mining and energy
113 Functional Managers and Directors
1131 Financial managers and directors
1132 Marketing and sales directors
1133 Purchasing managers and directors
1134 Advertising and public relations directors
1135 Human resource managers and directors
1136 Information technology and telecommunications directors
1139 Functional managers and directors n.e.c.

Obviously, few people know the SOC codes for a given occupation. Therefore the LMI API allows users and developers to search for a SOC code based on a plain text search. The search will return a maximum of 30 items, sorted by relevance. Some search terms, for instance engineer, will return multiple replies as engineer has many minor and unit groups, others less so. Search does not work for qualifications, because they are usually very similar between similar jobs.

The SOC code represents the building block for applications based on LMI for All, allowing the querying of the different data sets. However, please note that in some cases we are not able to return data based on a four-figure code. This may be because of a lack of statistically reliable data at this level of disaggregation or because the data has been deemed potentially disclosive. In these cases we will return data based on a higher SOC level and this will be flagged within the API Explorer.

Data available from LMI for All

In addition to the Standard Occupational Classifications, the current version of the LMI for All database contains the following key data sets:

  1. Employment, projected employment and replacement demands from Working Futures
  2. Pay based on the Annual Survey of Hours and Earnings and the Labour Force Survey
  3. Hours based on the Annual Survey of Hours and Earnings
  4. Unemployment rates based on the Labour Force Survey
  5. Skills shortage vacancies based on the Employer Skills Survey
  6. Skills, Knowledge, Abilities and Interests based on the United States O*NET system mapped to UK SOC
  7. Current vacancies available from Findajob
  8. Higher education destinations data from the Higher Education Statistical Agency

Access to the LMI for All database is provided through an API.

The following section provides ideas about the potential use of each dataset along with the different dimensions of the data. For more detail on the datasets have a look at the Data documentation pages. This includes the description of the dataset and provenance, details of the owner / curator, known quality issues with the data, quality control processes, accuracy of data, frequency of updates and issues around disclosure and confidentiality.

Employment data

Employment data is central to understanding opportunities for employment in different occupations. Whilst employment in some occupations will be broadly similar in different UK counties and regions in other occupations it will vary greatly on a geographical basis. Even though employment in some occupations may be declining overall, there will still be substantial opportunities in the future due to the sheer size of an occupation (such as engineering) and to replacement demand caused by retirement and people leaving an industry. The database not only contains historical information of actual employment levels but provides projections for the next 10 years.

The database provides very detailed information on the labour market which covers the following dimensions:

  • detailed SOC2010 4 digit occupational categories;
  • highest qualification held (thus if a person has several different occupation, only the one at the highest level will be recorded);
  • industry;
  • countries and English regions within the UK;
  • gender; and
  • employment status (full-time and part-time employees or self-employment). As well as historical employment levels this also includes projected Employment levels and estimates of Replacement Demands (which is a measure of job openings likely to arise over a selected period because of people currently employed retiring or leaving for other reasons).

Pay

Young people exploring possible future careers or unemployed people seeking work want to know how much they might earn in any occupation. The LMI for All API provides detailed pay data for different occupations covering the period 2012 onwards. Pay data are available for the following dimensions:

  • detailed SOC2010 4 digit occupational categories;
  • highest qualification held;
  • industry;
  • countries and English regions within the UK;
  • gender;
  • age; and
  • employment status (full-time and part-time employees or self-employment).

Both the mean and median pay level is available.

Hours worked

This dataset provides details of the mean number of hours worked in each occupation. Research shows that this is a key aspect of conditions of employment that people who are considering a career want information about.  Once more the data covers the period 2012 onwards. Unlike the employment and the pay and earnings datasets this one does not cover qualifications.

Hours data are available for the following dimensions:

  • detailed SOC2010 4 digit occupational categories;
  • industry;
  • countries and English regions within the UK;
  • gender; and
  • employment status (full-time and part-time employees or self-employment).

Unemployment rates

Detailed unemployment rates for different occupations in different UK countries and regions may be useful for understanding trends in employment, especially for planners. It also provides a direct indicator for job prospects in an occupation and the likelihood or otherwise of becoming unemployed. Unemployment rates are available for the following dimensions:

  • detailed SOC2010 4 digit occupational categories;
  • industry;
  • countries and English regions within the UK;
  • gender; and
  • employment status (full-time and part-time employees or self-employment).

However, small sample sizes mean that many of the more detailed results cannot be generated or published because of the lack of statistical robustness. This particularly applies to occupations with smaller employment numbers.

Number of vacancies

As with unemployment rates, vacancy rates are another useful source of information or gauging demand in particular occupations. The data on vacancy rates provided through he LMI for All API is based on the Employer Skills Survey and focuses on skills shortage vacancies for 2011, 2013, 2015 and 2017. This is based on a business survey which asks employers whether they have vacancies, whether they are hard to fill and whether they are hard to fill due a lack of applicants with the required skills.

Skills, knowledge, abilities and interests

The US O*Net system provides a broad range of data focused on the skills, abilities and interests associated with different occupations. The data is based on a large scale annual survey. This data can be used in a number of different ways for instance allowing users to explore how their particular skills and interests match different occupational demands. Although, the US has a different occupational classification system to the UK, the data is linked to SOC2010 4 digit categories by a mapping that links each SOC2010 4 digit occupation with one or more US occupations.

Occupational descriptions

Occupational descriptions include the content of each occupations, progression routes and often qualifications required. The LMI for All database provides a detailed structure of each SOC2010 occupation together with descriptions for each occupation. These data are from the ONS Standard Occupational Classification database.

Current vacancies

Obviously job seekers using data from LMI for All, will want to know what vacancies are currently available in their area. The data to provide this is currently taken from the Government’s Findajob bulletin board service. Although Findajob uses a different classification system, a keyword search suggests vacancies classified under SOC2010.

Higher Education

Many young people want to explore what jobs they could do with a degree in a particular subject. Unfortunately there is no way of directly linking course subjects to the SOC2010 classification system. However, we can provide data showing the destination of graduates six months after completing their course, in terms of the occupation that they are working in at that point.  This gives a useful indication of the course subjects that are relevant to particular occupations.  This data is from the Higher Education Statistical Agency (HESA).

Updates

LMI for All data is regularly updated. Full details of the update schedule can be found here.

License

All datasets included in the online portal are licensed under the Open Government License and should be used only under the conditions specified within the license. Details of the Open Government License can be found at here. These terms are compatible with the Creative Commons Attribution License 4.0 and the Open Data Commons Attribution License, both of which license copyright and database rights.

Full details of the Terms and Conditions of use can be found on the LMI for All website.

Mash-ups: Combining LMI for All data with other datasets

Of course LMI for All data does not cover everything. To extend the scope of applications based on LMI for All, we encourage developers to mash our data with other sources of data.

One approach is to link the data through SOC2010. An example of this is icould who provide over 1,000 careers videos from a wide from a wide range of occupations to offer an inside view of their current job and a personal account of how they got there. The videos have been classified against SOC2010 and using data from LMI for All under each video they are able to provide job information in the form of details of average salary, qualifications, skills, past and future employment levels, etc.

In other cases LMI for All data may be supplemented by data which is not necessarily classified to SOC2010. An app developed for the Department of Works and Pensions not only provides a dashboard to data from LMI for All, but also provides access to local authority labour market data from NOMIS.

Discussions with work coaches suggest that travel to work time and especially public transport is important for those seeking employment. This data is available from the National Public Transport Access Nodes (NaPTAN) database and could be mashed with LMI for All data.

And of course although LMI for All provides access to vacancies from Findajob, there are many other agencies providing job bulletin boards and some may be happy to have their data included in an application if it provides them with referrals.

Many applications will want some form of geographical mapping, using either data from Ordnance Survey or Open Street maps.

Other potential data sets could include:

Help

If you have any questions or require help please access the LMI for All Developers bulletin board. Talk about your apps, problems, bugs, the data and anything else to do with LMI for All here. Members of the project team are watching these discussions and will answer your questions.