Extended SOC 2020

Many Standard Occupational Classification (SOC) users reported that the existing four-digit structure is not detailed enough for their needs.

The extended SOC has addressed this need by creating an additional level of detail within SOC 2020. This level of the classification is termed "Sub Unit Group" (SUG) and comprises six digits. An initial version of the extended framework was released for publication in April 2021. Following ongoing development and refinement a final version of the framework is now available.

This final version shows an increase in the number of occupational unit groups from 412 within the four-digit structure to 1,369 at six digits.

Users of the SOC have told us that a more detailed SOC has potential to:

• give better understanding of labour market trends

• enable planning for future changes to labour markets

• improve careers advice services for individuals

• provide a universal product enabling a standardised approach to occupational groupings

Longer-term goals include the adoption of the extended framework into questionnaire design and statistical production. Supporting materials have been developed to aid this ambition. These include a version of the SOC 2020 Index of job titles matched to the extended six-digit SOC. This enables the look-up of over 30,000 UK occupations to their corresponding six-digit identifier. This represents the first step in enabling the automatic matching of data at the extended level.

SUG descriptions have also been developed. These will add clarity around the types of occupation included within a specific SUG and will aid both the automatic and manual matching of data.

Extended SOC2020 structure and descriptions (excel) 18-05-23 (341.4 kB xlsx)

Extended SOC2020 structure and descriptions (csv) 18-05-23 (152.8 kB zip)

SOC2020 Volume2 the coding index (excel) 22-02-24 (4.2 MB xlsx)

SOC2020 Volume2 the coding index (csv) 22-04-24 (1.3 MB zip)

The development of the extended SOC has been informed by several workstreams, while activities were overseen by a steering group chaired by Sir John Holman.

User engagement

Throughout 2019, the SOC Extension team worked with Professor Peter Elias from the University of Warwick’s Institute for Employment Research (IER) to conduct an extensive engagement exercise with main stakeholders. A total of 19 meetings were held with organisations identified as having significant knowledge of and interest in the SOC and the extension project.

Participants included:

• Careers Wales

• Department for Culture, Media and Sport (DCMS)

• Department for Education (DfE)

• Department for Work and Pensions (DWP)

• Health and Safety Executive (HSE)

• Higher Education Statistics Agency (HESA)

• Higher Education Careers Services Unit (HECSU)

• Home Office

• Learning and Work Institute

• National Audit Office

• Nesta

• Northern Ireland Statistical and Research Agency (NISRA)

• NHS Digital

• Office for Students

• Scottish Government

• Skills Development Scotland

• The Gatsby Foundation

• University College London (UCL)

• Welsh Government

Online survey of stakeholders

A stakeholder database of over 1,000 stakeholders was developed, representative of Minor Groups across the classification. Stakeholder mapping identified the degree of awareness and level of interest from stakeholders, which enabled targeted engagement across the database.

The survey tool was hosted on the ONS Consultation Hub for 16 weeks between June and September 2019. The survey asked:

• whether and how often respondents used the SOC

• whether the current version of SOC was detailed enough for their needs

• whether they felt their occupational area was sufficiently represented within SOC

• which areas would benefit from more detail?

• for examples of the additional detail required

• for examples of job titles and descriptions within respondents’ occupational area

A total of 183 responses were received by the electronic survey, which included representation across all nine Major Groups, 92% of Sub Major Groups and 79% of Minor Groups.

There was significant support for adding greater detail to the SOC, with around two-thirds of stakeholders indicating that there were areas where they would like to see greater detail added. Specific examples of where respondents requested greater visibility and detail within the classification included:

• cyber security

• event planning

• the green economy

• engineering

• the craft industry

Desk-based research

The extension has been further informed by a range of alternative classifications and secondary data sources.

Data

Data from Census 2011, Census 2021, Labour Force Survey and DLHE (Destination of Leavers from Higher Education) provided evidence of reported job titles. The use of these data helped identify quantitatively where disaggregation may be necessary and achievable.

International classifications

International classifications such as O*net (United States) and the European Skills Competencies and Qualifications Framework (ESCO) were used for cross-reference of occupational groups.

Bespoke classifications

Bespoke classifications adopted by stakeholders such as HESA, Careers Wales and NHS Digital were used to inform where user demand lay. Additional consideration was given to classifications developed by private companies such as Burning Glass, who collate job adverts from several thousand sources such as employment websites, specialised job portals and company websites into a single database.

Online research

Online research, including career websites and job vacancy portals, were used to understand the different skills and duties involved in job titles and to determine whether they were sufficiently distinct from others within the unit group to warrant disaggregation.

Quality assurance

The combined output of both primary and secondary research was used to objectively inform the development of the SOC extension structure.

Draft SUGs were developed by a research team within the Classifications team using the evidence available from the sources outlined in the previous section. These were then quality assured by colleagues with additional research being carried out as necessary until agreement was reached that the breakdown was appropriate and supported by the available evidence.

Further research and stakeholder consultation was carried out throughout 2022 to finalise the number of SUGs. The number of SUGs decreased from 1,463 to 1,369.

Principles of development

Following stakeholder feedback, the production of the extended SOC has been informed and guided by the following principles.

Each SUG within the classification should consist of a distinct and identifiable set of jobs. Considerable similarity must exist in terms of skill level and skill specialisation of the component tasks, which define each job within the SUG. The exception is for the ****/99 categories, which are defined as catch-all categories.

The ****/99 SUGs consist of two types of jobs:

• those which do not fit within any other SUG within the unit group, but are not yet sufficiently well established to constitute a SUG in their own right; these can be denoted by the phrase “not elsewhere classified” (n.e.c.)

• those which are not well defined so that clear allocation to a SUG within the unit group is possible; these could be denoted by the phrase “not otherwise specified” (n.o.s.)

• a SUG should be recognisable by its name and names of SUGs should not be ambiguous; the nomenclature of a SUG should reflect the name of the unit group

• there should be a good balance between the need for more detail in the classification in all areas, not just in areas where the identification of a SUG appears straightforward or demand has been identified from a specific SOC user

Automatic matching of data

The Office for National Statistics (ONS) is internally developing a matching tool. This tool was successfully used to assign Census 2021 data to four-digit SOC. Work is ongoing to maximise the match rates of test data by adapting a range of variables. Once complete the tool will be in a position to be used with live data.

Viability of statistical production at six digits

The feasibility of statistical production at the extended level is being explored internally. Statistical production relies on the adoption of the extended SOC to existing data sources. Potential challenges to this exist in terms of automatic match rates, data quality and sample size restrictions.

Despite this, a move towards matching data to the extended level of the SOC and subsequently, statistical production, remains in scope. Further work is required to identify how the challenges may be overcome.

Themed markers

The ability of a more granular SOC to identify science, technology, engineering and mathematics, plus medicine and health (STEM+MH) occupations from non-STEM+MH, was viewed as an important benefit from the extended SOC. The ONS has proposed adding further benefit by applying a marker to the framework, enabling STEM+MH occupation groups to be clearly identified and clustered together. Aggregation of this kind has the potential to enable statistical production in instances where the low numbers at a SUG level would be restrictive.

Stakeholder engagement also revealed several additional themed areas of interest including the creative industries, digital economy and the green economy. Research is ongoing with an aspiration that this could eventually lead to a suite of markers allowing SUGs to be aggregated by themes to produce statistics.

Stakeholders have also shown an interest in linking skills to the extended SOC SUGs.

Endorsement of the extended SOC

The extended SOC will be evaluated against the Taxonomy Best Practice Framework, which provides principles against which taxonomies suitable for use across government can be evaluated.