Question to the Cabinet Office:

To ask His Majesty's Government, further to the Written Answer by Lord Livermore on 17 February (HL14181), whether Office for National Statistics has assessed the feasibility and statistical acceptability of aggregating multiple Standard Occupational Classification codes that individually fall below publication thresholds into higher-level sector groupings for publication purposes; and if so, what conclusions were reached.

The information requested falls under the remit of the UK Statistics Authority.

Please see the letter attached from the Permanent Secretary for the Office of National Statistics.

The Rt Hon. the Lord Freybeg

House of Lords

London

SW1A 0PW

02 March 2026

Dear Lord Freyberg,

As Permanent Secretary of the Office for National Statistics (ONS), I am responding to your Parliamentary Question asking a) further to the Written Answer by Lord Livermore on 17 February (HL14181), whether the Office for National Statistics has assessed the feasibility and statistical acceptability of aggregating multiple Standard Occupational Classification codes that individually fall below publication thresholds into higher-level sector groupings for publication purposes; and if so, what conclusions were reached (HL14677); and b) what statistical disclosure control techniques, including aggregation, rounding, banding or noise adjustment, are available to the Office for National Statistics to enable publication of sectorlevel occupational data derived from low-count Standard Occupational Classification categories (HL14679).

The Standard Occupational Classification (SOC) codes are routinely used for statistical outputs. SOC codes are designed with a hierarchical structure, ranging from 1-digit codes for the broadest categories, 2 = ‘Professional occupations’ for example, to 4-digit codes for the most detailed categories, 2451 = ‘Architects’ for example. The SOC ‘nested’ structure means multiple Standard Occupation Classification codes are aggregated into higher-level groupings by design. The use of these standard groupings allows for better comparisons over time and with other data sources. Statistical outputs that the ONS produces will use an appropriate level of detail of SOC codes, aiming to balance the need for detailed occupation information, with the possibility of smaller categories falling below the publication threshold. Further information about the SOC classification can be found on the ONS website. There is currently a consultation to update SOC2020 because of the continual evolution of occupations, and to ensure that SOC reflects significant changes in the labour market. Submissions to the consultation are live until 11 May 2026.

All the disclosure control techniques you reference in your question are available to use. Aggregation and banding are the most used methods to increase the number of contributors. The choice of disclosure method depends on the source of data, user needs, disclosure risk and other related factors. The disclosure control methods applied are always designed to maximise the usefulness of the data whilst protecting the confidentiality of the contributors.

Yours sincerely,

Darren Tierney

https://www.ons.gov.uk/methodology/classificationsandstandards/standardoccupationalclassificationsoc/soc2020/soc2020volume1structureanddescriptionsofunitgroups

https://consultations.ons.gov.uk/external-affairs/soc2020-revision-consultation/