Recommendations

We have identified a number of actions that we would like ONS to take in response to our findings. These are set out below.

To ensure future population statistics are based on sound methods and suitable data, ONS’s population estimates and projections team needs to:

• Investigate the root and scale of the issue associated with cities with large student populations and communicate its findings publicly, to support the appropriate use of the existing data.
• Use its partnership with experts to discuss the evidence provided to OSR in the review concerning the impact of assumptions being rolled forward.
• Assure itself and others that concerns raised regarding the current methods are considered throughout the development of its admin-based population estimates.

To enhance the transparency of developments concerning the quality of the statistics, ONS should:

• Integrate a more flexible and responsive approach to methodological changes in its design for admin-based population estimates. While we appreciate that there should not be adjustments made in response to every concern raised, ONS’s population estimates and projections team should work with its partners in local government, academia and across the devolved nations, so that changes are implemented in a more timely way.
• Share the insights it has gathered from the work it is doing to understand the changing nature of migration and population, as part of its transformation programme, so that users’ views inform the way this work is taken forward.

As ONS continues to develop its long-term plans for the future of migration data, ONS’s migration team should be open with users about its short-term solution to bridge the gap of migration data until the administrative data alternative is fit for purpose and ready to use.

To enhance its approach to quality assurance, ONS’s population estimates and projections team should:

• Collaborate with others to learn from best practice – for example learning from demographers and the Greater London Authority who produce their own estimates and projections.
• Incorporate local insight and evidence as part of its deep dives and investigations into issues.
• Run sensitivity analyses to accompany the existing estimates and explain to users how these analyses should be interpreted.

To support users’ understanding of the uncertainty associated with these statistics, ONS’s population estimates and projections team should:

• Research and implement additional ways to communicate the uncertainty around the population estimates and projections, beyond the use of confidence intervals and variant projections.
• Provide more specific guidance on interpreting the levels of uncertainty associated with the statistics, to help users understand the appropriate use of the statistics for short-term planning compared with longer-term planning.

To maximise the use of the variant projections, ONS’s population estimates and projections team should expand on the support it gives users to illustrate where the use of these alternative projections may be beneficial and develop case studies of where they have been used in practice.

To ensure the statistics remain relevant to users, ONS’s population estimates and projections team should:

• Take a more open and constructive approach to responding to user feedback by improving its complaints procedure and viewing challenge as an opportunity to improve the statistics and outputs. A fully open approach will help ONS demonstrate its commitment to user engagement and ensure a range of perspectives are fed into the development of the statistics.
• Reflect and learn from its experience of challenging user engagement and identify potential solutions and best practice from the User Engagement Strategy for Statistics.

To increase the public value of these statistics and support their use, ONS’s population estimates and projections team should:

• Carry out user engagement to understand who is using the data and for what purposes. Through this, it should promote the appropriate use of the data.
• Collaborate with experts to frame the statistics for different audiences and scenarios, presenting appropriate use cases of the data.

Next Steps