Present   Members   

Dame Moira Gibb (Chair)   

Rob Bumpstead  

Vanessa Cuthill   

Colin Godbold  

Isabel Nisbet   

Marion Oswald   

Emma Uprichard 

UK Statistics Authority    

Daniel Towler  

Ross Young 

Rhys Nadin (for item 7)  

Karen White (for item 7)  

Office for National Statistics   

Anita Arif (for item 4) 

Emma Parker (for item 5) 

Maria Tortoriello (for item 5) 

Office for Statistics Regulation   

Mary Cowan (for item 6) 

Other   

Kathryn Helliwell, Welsh Government (for item 2) 

Ffion Lloyd-Williams, Welsh Government (for item 2) 

Polina Obolenskaya, London School of Economics (for item 3) 

Polly Vizard, London School of Economics (for item 3) 

 Apologies   

Stephen Balchin   

Lily O’Flynn 

Simon Whitworth 

1. Minutes and matters arising from the previous meeting

1.1       The Chair welcomed members to the 24^th meeting of the National Statistician’s Data Ethics Advisory Committee (NSDEC).

1.2       Members approved the minutes from the previous meeting.

1.3       Daniel Towler updated the Committee with progress on actions from previous meetings and correspondence.

2. EU Settled Status Scheme (EUSS) Data Linkage Project. NSDEC(21)07

2.1       Ffion Lloyd-Williams and Kathryn Halliwell presented this project which is an ADR Wales initiative that aims to link Home Office EU Settled Status (EUSS) data with administrative and health data held in the SAIL Databank. The data linkage is designed to enable researchers and policymakers to better understand the experiences of EU citizens with Settled Status, at an aggregated level, and therefore to adapt policies and public services to this population group.

2.2       The following points were raised in the discussion:

• The Committee wished to further understand the specific research purposes/aims from the finalised dataset that require such linkages to be undertaken;
• Additional assurance that this data linkage project is being driven out of research demand to answer vital public good questions was requested. The Committee requested that the creation of this dataset was subject to regular review to consider research need and dataset suitability to ensure that the linked dataset is created to meet specific research purposes;
• The Committee wanted to understand what guarantees are in place to ensure that this dataset will continue to be used solely for research and statistical purposes in the future;
• The sensitivities of timing should also be considered due to the imminent deadline for EUSS application submissions. The researchers should ensure that there isn’t a perception that the project may develop into identifying those without settled status;
• More detail regarding the health data proposed to be included in the analysis would be useful given the sensitivity of such data;
• The Committee suggested it may be useful for the application to be amended to make it clear that the research is specifically about this population group and not the whole UK population;
• It was suggested that an equality impact assessment would be valuable;
• The Committee recommended that the researchers continue to carefully consider the sensitivities of this research. For example, the researchers should ensure that those with EU settled status are not necessarily considered as one homogenous group given the differences that exist within this group, and;
• The Committee emphasised that measures should be considered to ensure those excluded from EUSS data are not disadvantaged.

2.3       The Committee made clear their support for data linkage, but reiterated the need for a clear purpose and benefit for doing so.

2.4       Action: Researchers to work with the data ethics team to provide assurance of the above points, and present future uses of the data at a future NSDEC meeting.

3. Shaping, testing and demonstrating the value of the Growing up in England wave 1 dataset: Roma, Gypsy and Traveller children case study. NSDEC(21)08

3.1       Polly Vizard and Polina Obolenskaya (LSE) presented a project which sets out the first use of Wave 1 of the Growing up in England (GUIE) dataset, using linked 2011 Census (ONS) and the Department for Education’s ‘All Education Data for England’ data (AEDE). The project aims to shape, test and demonstrate the value of the newly linked wave 1 of the ‘Growing Up in England’ (GUIE) dataset to provide new quantitative evidence focussing on the educational experiences of Roma, Gypsy and Traveller (RGT) young people and children as a case study.

3.2       The Committee commented on the development of this linked dataset, which was initially called ‘Data for Children’ in 2019. This application stated that research using this data via the Research Accreditation Panel should be have the ethics appropriately considered.

3.3       This project is being reviewed by NSDEC prior to submission to the UKSA Research Accreditation Panel for accreditation under the Research strand of the Digital Economy Act in due course.

3.4       The following points were raised in the discussion:

• The Committee requested further assurance that the researchers are engaging with the RGT community, and that the RGT community supports the public good of this work;
• The Committee recommended that the researchers consider that the ‘education equals progress’ assumption may not necessarily be shared by all;
• The application would benefit from more detail on the outputs that will be produced, how they will be published, and whether there are any potential consequences for the RGT community and consequently how these will be mitigated;
• The Committee questioned whether those children who may be absent from education records for one year, but later reappear, would be excluded from the dataset;
• The Committee suggested that there is consideration of missing data, for example, as a result of self-identification of key characteristics in both the census and the All Education Data for England data, and also missing data associated with exclusions and periods of absence from school, and;
• As this project is not confined to children, the Committee suggested that there is further consideration of the rights of adults within the study.

3.5       The Committee acknowledged the thoroughness of the application and appreciated the value of the study.

3.6       Action: LSE researchers to provide the secretariat with an updated application reflecting the considerations and suggestions of the Committee.

4. Attitudes to Data Sharing and Administrative Data

4.1       Anita Arif, in the Strategic Communications team at the ONS, presented ONS research which sought to understand public attitudes to data sharing, administrative data, and administrative data linkage.

4.2       The following points were raised in the discussion:

• The Committee suggested it may be useful to further consider public views on the security of data;
• The Committee recognised the similarity of the findings to a number of previous studies, and felt that the research might benefit from providing additional clarity on those elements of this research which have produced novel insights in this area, and those which reiterate findings from previous studies on public attitudes to data sharing;
• It was proposed that the ONS may benefit from collaborating with the UKSA Centre for Applied Data Ethics to produce further guidance on the issue, particularly in the areas of public attitudes, data sharing and data ethics, and;
• The Committee suggested that it would be useful to publish these findings to the wider public at an appropriate point.

5. Telematching Data

5.1       Maria Tortoriello, in the Social Survey Transformation team at the ONS, presented an overview of the ONS’s increasing use of telematching data in social survey collection, as new methods of data collection are required to boost response rates during the pandemic. ONS sought NSDEC’s advice on the proposed scaling-up of use of telematched data, which is a chargeable service offered by commercial organisations whereby a telephone number is appended to a sampled address.

5.2       The following points were raised in the discussion:

• The Committee stressed the importance of ensuring that the commercial organisation(s) supplying the telephone numbers are fully compliant with the ethical standards of the UK Statistics Authority, should ONS decide to increase its use of telematched data;
• It was emphasised that the ONS must be careful to avoid a perception of engaging in ‘cold-calling’, and uphold their safeguarding responsibilities in the event of contacting vulnerable groups;
• ONS should consider whether the increase of 4.5% in responses due to telematching has been impacted by the Covid-19 pandemic due to respondents being at home more, and factor in the potential for success in improving survey response rates while deciding on the future use of telematching;
• The Committee suggested that the ONS should liaise with UKSA/ONS Data Protection Officer and team to assess whether the Information Commissioner’s Office needs to be consulted; and,
• The Committee agreed that it may be useful to understand the demographics of the success rates of telematching to understand the sustainability of the method.

5.3       The Committee agreed that it is of utmost importance that the ethical risk is proportionate to the potential benefit.

5.4       Action: Researchers to consider the Committee’s advice when deciding whether to move forward with the increased use of telematched data. Should this policy move ahead, the researchers are advised to work with the secretariat during the tendering process, and keep the Committee updated of progress.

6. Defining the Public Good in Applications to Access Public Data

6.1       Mary Cowan, of the Office for Statistics Regulation (OSR), presented OSR’s Research Programme work  to understand how the public good of statistics is conceptualised by researchers and analysts applying for access to public data, to enable the UKSA to further understand what ‘public good’ means. This work will feed into a report that will be published by OSR on understanding how researchers view the public good.

6.2       The following points were raised in the discussion:

• The Committee agreed on the importance of exploring what is meant by the public good to enable researchers to successfully apply it to their work and in their applications, and;
• It was suggested that the UKSA Centre for Applied Data Ethics could provide additional support and insights with the development of this work, given that public good is upheld in the UK Statistics Authority’s ethical principles.

6.3       The Committee acknowledged the clear importance and value of this study and its early findings and were very appreciative of the presentation. The Committee would welcome the opportunity to be updated on the progress of the research, to enable a more substantive discussion.

6.4       Action: Secretariat to work with the Office for Statistics Regulation to enable further discussion on the findings of this research at a future meeting.

7. Plan for Ethics Compliance Reviews

7.1       Rhys Nadin and Karen White, in the Data Protection Compliance team at the UK Statistics Authority, presented this item on ethics compliance reviews. This paper set out a forward plan of suggested ethics compliance reviews for the next four quarters. This paper also presented a data ethics compliance review on the Child Prevalence Survey Feasibility Study, undertaken by the ONS, which was considered by the Committee in 2019.

7.2       The review concluded that the research team has considered, acted upon, and provided sufficient progress updates on the recommendations that NSDEC provided in October 2019.

7.3       The Committee suggested that it would be useful to see more projects (beyond those led by ONS) included on the forward plan.

7.4       The Committee recognised the quality of the work and were grateful to the team for the assurance that NSDEC’s advice is being properly considered.

8. Any Other Business

8.1       The Chair reminded the Committee to log into the new collaboration platform to review project applications, recently operationalised, and agreed to at the February 2021 meeting.

8.2       The next meeting will be held on 30 June 2021.