Guidance

DBT statistics error policy

Updated 2 October 2024

This guidance documents the error policy (including revision policy) for statistics produced by the Department for Business and Trade (DBT), in line with practices recommended in the Code of Practice for Statistics, specifically setting out the department’s commitments on T2: Independent decision making and leadership, and Q3: Assured quality.

This guidance is designed to be used by DBT staff responsible for the production, use and dissemination of statistics, or in the handling of statistics. It contains a set of rules to ensure that all DBT officials use statistics effectively when making decisions in DBT and handle the statistics in a way that minimises risk to the DBT. This will be reviewed at least annually. The preceding version of this guidance is available from the National Archives.

1. DBT error policy for published statistics

This policy covers the production and publication of DBT official statistics, as well as any statistics that are used externally.

1.1 Scheduled revisions

A scheduled revision is where statistics are initially published as provisional, which means that they are subject to change. At a later scheduled date, as announced on the upcoming statistical releases on GOV.UK, the statistics will be revised and reissued.

Statistics published as provisional should be clearly marked by a superscript ‘p’ next to each provisional figure and include information on the reasons why the data is provisional.

Once data initially published as provisional is made final, the ‘p’ markers should be removed and any figures that have changed should be marked with an ‘r’. Information on the reasons why the data has been revised should be provided.

1.2 Corrections

Corrections to statistics will be made if errors are found or if figures are amended because of changes to methodology or definitions of the data collection.

Where an error has been found, the Chief Statistician and/or Head of Profession for Statistics should be notified as soon possible. They should be given details of the error, its scale and proposed corrective actions.

The Chief Statistician and/or Head of Profession for Statistics will decide if a major or minor error has occurred and advise the team responsible for the publication of any required corrective actions.

Major errors are errors which have a significant impact in the use made of the data, for example where the overarching message changes. These errors should be corrected as soon as possible by reissuing the statistics with the corrected figures. In such circumstances key users and/or ministers may be alerted.

Minor errors are all other errors which are not classified as major errors. These can be corrected in slower time as appropriate (always within 12 months) and can be timed to coincide with, but are not restricted to, the release of new editions of statistical products.

In cases where errors are found in time series that appear in more than one edition, the errors should be corrected in the most recent edition only. Older editions need not be corrected.

Where changes are made to methodology or to definitions in a time series, corrections to the historical data in the time series should be considered to bring it in line with the new methodology or definitions. This may not be possible and depends on the data availability and the amount of work involved. The requirements of the users should be considered when making such decisions.

If an error is identified but corrected figures are not yet available due to a need for further investigation, a note should be added to the statistics to explain the situation, making it clear the figures are subject to change.

Corrections should be shown by the addition of a superscript ‘r’ next to figures that have been corrected. This should be accompanied by a note explaining the reason for and the scale of the correction. Corrections will only be marked in the first edition of each publication that the corrected figures appear in. If the figures then appear again in future editions. they will not be marked as corrected.

The report should state when it was first published and when it was revised.

In cases where errors are corrected in cumulative spreadsheets, that is, where the latest data is added to a spreadsheet, which then replaces the previous edition on GOV.UK:

• in the first edition after revision, corrections should be shown by the addition of a superscript ‘r’ next to figures that have been corrected. This should be accompanied by a note explaining the reason for and the scale of the correction

• in subsequent editions, the ‘r’ marker and original note should be removed from the spreadsheet and replaced by a note explaining what was revised, the reason for and scale of the correction

• the old uncorrected publication will be removed but should be filed and available on request

• steps should be taken to determine what corrective actions/training/communications are needed to prevent similar errors from occurring in the future. The emphasis is on learning from mistakes and improvement

2. DBT error policy for Parliamentary questions

There should be no instance in which a Parliamentary question (PQ) response is expected to be revised at a later date. If data is not currently held in a format that enables a sufficient response to the PQ, an explanation should be given for why an answer could not be provided.

If a PQ answer is found to contain an error, a correction needs to be made. The corrected answer will appear on the Parliamentary website alongside the earlier response in a ‘correction to the official report’.

3. Annex A: DBT publication notices

The following disclaimer is included in DBT’s official statistic publications:

3.1 Disclaimer

Whereas every effort has been made to ensure that the information in this document is accurate, the Department for Business and Trade does not accept liability for any errors, omissions or misleading statements, and no warranty is given or responsibility accepted as to the standing of any individual, firm, company or other organisation mentioned.

The following examples are the published revision policies of DBT’s Official and National statistics (available on gov.uk).

3.2 Business population estimates (BPE) – revisions policy

When re-weighted (for updated population estimates) ONS Labour Force Survey (LFS) data is made available, headline estimates of the number of private sector businesses in the UK for the relevant years will be revised in case the re-weighting significantly affects the estimates. Updated estimates by size band will be provided in figure 3 and table B of the statistical release and table 25 of the detailed data tables, to reflect the impact on the unregistered business population estimate.

Since 2018, no re-weighted data has been published. The quantification of uncertainty in estimates of unregistered business numbers has shown the effects of re-weighting have been relatively minor compared to the uncertainty inherent in the estimates. Until this changes, there are no plans to re-weight data.

Revisions of previously released estimates will also be carried out where new sources of information become available that improve the robustness of the BPE. Any future substantial methodological changes of this nature will be consulted on and pre-announced before being implemented.

3.3 Longitudinal Small Business Survey – revisions policy

It is not expected that historical data will be revised. Where any minor errors are uncovered, the department will update in the next annual publication (see statistics release calendar). Major errors or revisions will be made as soon as possible with reasons provided.

Any future substantial methodological changes will be consulted on.

3.4 UK Innovation Survey – revisions policy

It is not expected that historical data will be revised. Where any minor errors are uncovered the department will update in the next annual publication (see statistics release calendar). Major errors or revisions will be made as soon as possible with reasons provided.

3.5 Building materials and components: monthly statistics – revisions policy

Data can be revised for a number of reasons, including receipt of more data, methodology changes and corrections.

Data subject to revisions are published as provisional and marked [p], except for seasonally adjusted values in editions of publication published from February 2024 onwards, where relevant information is provided in footnotes. The period for which data are provisional varies across material types and is indicated in the table below. When provisional data becomes final, [p] marks are removed from tables.

For unscheduled revisions, published from February 2024 onwards, no marker is added to tables, though in case of substantial revisions a note is added at the top of each table in the edition of the publication in which the revision was first added.

For tables 9-12 of this publication, covering data for bricks and concrete blocks, data is often subject to revisions in the 3 months after it is initially obtained. These revisions are often due to respondents providing additional data or revising the data they provided before. As a result, the latest 3 months of data in these tables are provided as provisional. Any significant revisions to data in these tables will be reported in footnotes and the ‘Contents’ section. These changes were implemented in the August 2024 edition of the publication onwards. Prior to this, bricks data was considered provisional for 1 month, while concrete blocks data was finalised at the end of each quarter.

Revision schedule

Data series	Revision timing after the publication of provisional information
Table 1: Construction Material Price Indices	3 months
Tables 2 and 3: Price Indices of Construction Materials	The latest 5 months’ data are subject to revisions
Table 4: Sales of Sand and Gravel in Great Britain	1 month
Tables 9 - 12: Bricks and Concrete Blocks - Production, Deliveries and Stocks	The latest 3 months of data are regularly revised and are considered provisional.
Table 13 and 14: Value of Overseas Trade in Selected Materials and Components for Constructional Use	Up to 5 quarters

Users should download the latest edition of the publication to ensure they have the most up-to-date data.

If minor errors are uncovered, the department will update in the next publication (see statistics release calendar). Major error corrections or revisions will be made as soon as possible with reasons provided.

Any future substantial methodological changes will be consulted on.

3.6 Trade union membership statistics – revisions policy

This is an annual publication using data from 3 sources: the Labour Force Survey, the Annual Survey of Hours and Earnings and the trade unions’ annual returns to the Certification Officer. Where ONS revise or re-weight their survey data, these changes will be reflected in the next annual publication. The trade unions do not normally revise their membership data in annual returns. Where there are definitional or methodological changes, for instance with the forthcoming transformed LFS, they will be described in the publications prior and subsequent to the change (where practicable).

3.7 Core statistics book – revisions policy

On occasions, previously published data will need to be revised due to changes to source data or correcting of errors. Major revisions will be made as soon as possible while smaller revisions will be made at the time of the next scheduled release (see the statistics release calendar). Additionally, some data will be provisional and subject to future revisions. Where a large revision has taken place, reasons will be provided.

3.8 UK trade in numbers – reporting methodology

There is a variety of sources on trade and foreign direct investment (FDI) statistics. Estimates may differ between sources due to data revisions and differences in coverage, currency and methodology. The figures used in the pocketbook are the very latest available data at the time of publication. To be able to provide this information, a variety of sources have been used.

While these monthly/quarterly ONS releases are more timely, they are not fully consistent with geographical breakdowns. This means that in some instances, figures for the same metric (for example, total UK exports) may differ between sources.

There are also often discrepancies in the data published by the UK and international organisations such as United Nations Conference on Trade and Development (UNCTAD) and Organisation for Economic Co-operation and Development (OECD). These are mainly due to exchange rate conversions, delays in processing revisions and data adjustments applied by international compilers.

Due to some sources updating frequently, monthly in some cases, figures will be revised in future and therefore the figures presented in the pocketbook are an accurate snapshot at the time of publication. We encourage readers to look up and use the latest sources available.

3.9 Strategic export controls statistics – revisions policy

DBT publishes quarterly statistics on Strategic Export Controls covering export licensing decisions made by the government. This document sets out a revisions policy made by the Export Control Organisation (ECO) to revise or correct the publication in accordance with the UK Statistics Authority’s Code of Practice for Statistics.

Causes of revisions and policy for handling each type

ECO has identified 3 possible causes of revisions to published data: corrections, receipt of more data or data changes due to amendments and changes to data due to methodological changes.

Although ECO has robust procedures in place to minimise errors in the statistical systems and processes, errors may occasionally occur. If errors are found in the published statistics, corrections will be promptly made together with a correction notice to alert users.

Receipt of more data (including due to amendments to licences)

There are 2 main reasons for such changes: a licence issued during the period may have been revoked, for example because of the imposition of trade sanctions or an arms embargo; or a decision during the period to refuse a licence might be overturned because the applicant later appealed successfully. Additionally, there could also be a licence conversion whereby licences are converted, resulting in a change to the licence.

Each set of data tables in the quarterly strategic export controls statistics will revise data, back to 2008, if it has been changed for either of these reasons. The ‘r’ flag will be used where figures change between quarterly official statistics publications. As such, revisions are relative to the previously published data.

Each set of PDF country pivot reports in the quarterly strategic export controls statistics show revisions differently. The ‘Rv’ flag is placed against case cummaries that were issued and then revoked in the same reporting period. The ‘A’ flag is placed against case summaries that (some or all of which) were issued on appeal, following an initial refusal. Historical licensing data is updated where changes to licences have been made (for example, due to amendments) but these are not explicitly flagged.

Strategic export controls statistics are derived from the government’s licence management systems. Changes to underlying data collection methodology will cause revisions. These include changes in statistical methods, definitions, classification and geographic coverage.

ECO will, where possible, consult with users when major methodology changes are planned, and consider the options for maintaining a consistent time-series, including any revisions to previously published time-series. Revised data will be accompanied with an ‘r’ and footnotes explaining changes made to the methodology which has caused the data to be revised.

3.10 UK defence and security export statistics – revisions policy

All figures are subject to revisions due to definitional and methodological changes that will be notified to users via the GOV.UK website.

The analysis and statistics team, in the operations team at UK Defence and Security Exports, do not make adjustments for subsequent cancelled orders, unless they are cancelled before the figures are published. However, if we notice an error or come across additional information during the course of the year, that has substantial impact on the meaning of the figures, we will update and republish as soon as possible.

3.11 Great.gov.uk performance dashboard – revisions policy

It is not expected that historical data will be revised. Where any discrepancy is later uncovered in a subsequent month, the department will update in the next month’s release and issue a note to explain the reasons for this and any effects on the overall statistical series. Revised figures will be marked with an ‘r’.

Google is responsible for classifying and calculating the data on this dashboard provided by Google Analytics. You can read Google’s online guides for more information.