Population and International Migration Statistics Revisions Policy

This revisions policy is approved by the Office for National Statistics’ (ONS) Centre for International Migration and Centre for Ageing and Demography, and by the devolved administrations members of the UK Census and Population Statistics Strategic Group.

The policy relates to population and international migration statistics produced by the ONS. It also relates to any equivalent statistics published by the devolved administrations. This is to ensure consistency in the handling of revisions relating to UK-wide population and international migration statistics.

The National Records of Scotland (NRS) and the Northern Ireland Statistics and Research Agency (NISRA) produce and publish the population and migration statistics for Scotland and Northern Ireland, respectively. They may publish separate revisions policies on their own websites.

This policy has been developed in line with the relevant statistical regulatory framework. The Code of Practice for Statistics states that: “Producers of statistics and data should explain clearly how they assure themselves that statistics and data are accurate, reliable, coherent and timely”.

In the fourth practice of the quality pillar (Q3.4), it states that: “Scheduled revisions, or unscheduled corrections that result from errors, should be explained alongside the statistics, being clear on the scale, nature, cause and impact”.

In the ninth practice of the trustworthy pillar (T3.9), it states that: “Scheduled revisions or unscheduled corrections to the statistics and data should be released as soon as practicable. The changes should be handled transparently in line with a published policy”.

For more information, please see the UK Statistics Authority’s Code of Practice for Statistics.

The way that the Office for National Statistics (ONS) produces population and international migration statistics is being transformed. This is to better meet our users’ needs for a more timely and accurate understanding of the population on an ongoing basis. Central to this change in methodology is the use of administrative data, that is, collections of data primarily maintained for an operational purpose.

Our new methods make the best use of available data to produce the best possible statistics at a point in time. However, these provisional estimates are released with the expectation that they may be revised and updated as further and more complete data become available. This is particularly relevant for long-term international migration statistics, where actual behaviours observed in administrative data may not initially identify someone as a long-term migrant (as defined by the UN) until a period of 12 months has passed.

Our new methods are still at the experimental stage. Therefore, revisions of our statistics are also likely to occur in response to improvements in our methods.

Scheduled revisions are planned amendments to published statistics to improve quality. Statistical revisions are most commonly issued for four reasons, to:

• incorporate more complete or new source data that were not available at the time of the initial publication, or corrected reporting by respondents
• capture routine recalculations, such as revising provisional estimates and/or replacing them with final estimates
• replace projections about expected behaviour (such as whether an individual is expected to stay or leave the country for more than 12 months) with actual data
• reflect improved methodology such as changes in statistical methods or systems and changes in concepts, definitions and classifications; for instance, the models currently used to produce official estimates of international migration are under continuous development with improvements often applied to historical as well as new estimates

Major revisions and corrections (for any reason) will be accompanied at the time of publication by an explanation of their nature and extent. Updated back data will be provided if available and where they will add value.

The Office for National Statistics (ONS) will announce substantial methodological changes, their reasons and effects, with the release of statistics based on the new methods. In the case of key outputs, an information paper providing more detail on the methodological changes will be made available on the website.

In planning revisions relating to improved methodology, the ONS will carefully consider the balance between “bunching” revisions versus many successive revisions, in consultation with users. While users often prefer grouping changes together to avoid multiple revisions, this does lead to larger, more complex projects and difficulties in distinguishing separate effects.

Revisions to population and international migration statistics have wide-reaching implications and the need for them must be carefully assessed. There are implications for producers and direct users of population and international migration statistics and for the producers of other statistical series or outputs. There will always be a balance between the need for the most accurate and timely estimates and the need for stability.

The ONS should make users aware before, or alongside, releases of scheduled revisions and the magnitude of the effects, if possible. Unplanned revisions will be made in a transparent way.

Historically, the ONS has sought to avoid frequent revisions, to maximise the consistency of approach to our estimates over time.

The ONS will revise our population and international migration statistics where new source data or methods become available. For example, following new census data as part of our regular and historical population estimates rebasing process and/or new approaches to producing population and migration statistics based predominantly on administrative data.

The ONS are continuing to develop a transformed population and migration statistics system to provide more timely and frequent estimates. The demand for more timely statistics on the population may lead to an approach of provisional and final estimates being produced. This is to ensure that the latest available administrative data and methods are incorporated and the published estimates are of the highest possible quality.

Similarly, any important revisions to estimates of long-term international migration would affect the official population estimates and so require revisions accordingly. Where this occurs, the ONS will try to ensure that revisions are described prior to the production of our regular outputs.

Communication

The ONS will communicate the transformation of population and international migration statistics to key stakeholders as these develop, to ensure transparency. These changes may potentially lead to revisions. For the latest, see our Population and migration statistics system transformation – overview article, published 25 June 2021.

The ONS will introduce new methods where there is clear evidence that new or alternative methods are acceptable to users and of higher quality taking into account the balance of accuracy and timeliness.

The ONS is committed to updating and consulting users through online and face-to-face user groups during the research phase. This may last for several months or years before a new method is implemented.

When statistics are released using existing methods under review in long-term improvement projects, information accompanying their release will clearly reference the ongoing research and timescales.

When implementation of new methods is confirmed, substantial methodological changes will always be announced before or alongside the release of statistics. This will be based on new methods or classifications, together with an indication of the likely effects.

Where data allow, a revised back series for consistency over time will be provided. The ONS will aim to provide this back series alongside the first release of a new estimate based on the new method. However, when there is strong evidence that a new method is of higher quality, it may be used for the most recent reference period. This will be before a comparable back series is available. There will be an element of judgement in this, but user needs will be prioritised.

Unscheduled corrections are amendments made to published statistics in response to identified errors following their initial publication. The Office for National Statistics (ONS) will ensure that:

• if errors are discovered after publication, their impact will be investigated, and corrections made
• where errors are discovered and they are within the bounds of uncertainty, the ONS may not do a correction
• where a correction is made, a correction notice will be provided on the website as soon as possible to inform users and ensure transparency
• where corrections are assessed as having a major impact, they will be issued as soon as practicable
• where corrections are assessed as having a major impact, the ONS will make efforts to inform key stakeholders directly
• where corrections are assessed as having a minor impact, they will be incorporated in the next planned release if it is practical and cost effective to do so

Correction notices will be published on the release page and if the correction relates to a dataset page, they will also be included in the first tab of the spreadsheet.

If you require more information about this revisions policy or would like to feedback comments, please contact pop.info@ons.gov.uk. The Office for National Statistics (ONS) also invites comments on how users would like to be kept informed.