Response to the household projections for England compliance check
I am writing to update you on the progress made since your compliance check on household projections for England published on 17 December 2018.
We noted the letter was very positive about the way the team engaged with OSR, users and independent experts in the development of the 2016-based household projections for England, as well as the openness and transparency with which proposals were presented and responded to. Your letter requested an update from me about how we have addressed the recommendations made in the compliance check. Since receiving your letter, we have been working towards addressing these recommendations and I am now writing to provide the update you requested. I address each recommendation below, along with the actions we have taken and are planning to take in the future.
The team should review and refine the range of methods guidance published to enhance its accessibility to users.
We have reviewed the range of methods guidance published alongside the 2016-based household projections for England. We have updated the Quality Assurance of Administrative Data report about prison population data, to reflect their use in household projections; a link to the updated report will be included in updated methods guidance.
We are reviewing the household projections Quality and Methodology Information (QMI) report, to reduce duplication of information across documentation and provide a more accessible summary of the methods used. In addition, we have developed a flow diagram summarising our approach to producing household projections, which will also be included in updated methods guidance.
Our updated methods guidance will also include a clear summary of the main assumptions made in the production of household projections for England, how they compare to previous assumptions and why each assumption was made, building on the information made available in section 11 of the existing methodology guidance.
We plan to publish updated methods guidance for the 2016-based household projections in August 2019.
ONS should investigate Local Authority-level projection outliers and provide supplementary analysis and guidance about the key drivers for these results.
We have continued to monitor users’ feedback and respond to queries regarding the differences between the 2014-based and 2016-based household projections for England. In March 2019, we provided information and an opportunity for feedback and questions to local authorities as part of the Centre for Ageing and Demography stand at the Census local authority roadshows.
We are drafting analysis of local authority level projection outliers, based on the guidance and analysis which we have provided in response to customer queries since the 2016-based household projections were published. We intend to publish the results of this analysis in August 2019.
ONS should prioritise publishing additional ‘variant projections’ in the near future, in a way that is most helpful to users.
In February 2019, we carried out an informal user engagement exercise with stakeholders to identify user needs for variant household projections. Users were most interested in variant household projections based on the three variant subnational population projections (SNPPs) which were published on 9 April 2019. After these, users were most interested in a variant which projects household representative rates (HRRs) for the whole projection period (as opposed to the principal projection, which holds HRRs constant at 2021 levels for 2022 onwards). These are the four variants which we published on 16 May 2019.
Two other variants were identified in the survey as being particularly useful to users but were not included in the release on 16 May. These were: firstly, a variant that projects family units instead of households; and secondly, a variant that shows higher HRRs for 25- to 44-year-olds. These variants were not included in the May publication because further research is required to identify the most appropriate method for adjusting the assumptions underpinning them. We will continue to research the most appropriate methodology for producing them.
ONS should revisit and update published guidance about the coherence and comparability of projections produced for each of the four countries of the UK.
Updated guidance has been drafted in collaboration with the Welsh Government, National Records of Scotland (NRS) and the Northern Ireland Statistics and Research Agency (NISRA). The structure of the previous guidance report has been reviewed and updated to ensure that making comparisons between methods across countries is as straightforward as possible. Flow diagrams summarising the methods used to produce household projections for England, Wales, Scotland and Northern Ireland respectively will be included in the updated guidance. We intend to publish this guidance in August 2019.
We’ll keep you updated on our plans, and make sure to send you links to the various outputs that we publish over the summer. Please contact either Richard Pereira or I if you wish to discuss any aspects further.