Administrative Data Based Ethnicity Estimates (2022): Statistical Research

Background    

This publication provides an update to research on the Administrative Data Based Ethnicity Dataset (ABED) first published in February 2025 as part of the Administrative Data Based Population, Household and Ethnicity Estimates (2016-2022) - Statistical Research.  

The ABED uses administrative data and Scotland’s Census 2011 to produce estimates of the size of ethnic groups in Scotland. This publication discusses the 2022 ABED based on 20 March, the same date as Scotland’s Census Day. The ethnicity information from the administrative data is linked to Scotland’s Integrated Demographic Dataset (SIDD) containing the age, sex and location of individuals. The SIDD is used as the population base. 

This publication builds on the previous research which described the data sources and methods as well as the proportion of people which administrative data is able to establish an ethnicity for.  

It uses two methods, one involving administrative data only and the other using both administrative data and Census 2011 data. The research explores which ethnic groups are available in administrative data, the ethnic breakdown observed in the ABED, and compares the ABED estimates to Census 2022. 

NRS is continuing research to develop statistics produced using administrative data. Part of this development includes acquiring data from other providers including the Department for Work and Pensions (DWP) as well as exploring how the ABED can support Scotland’s Census 2031.  

Key Findings  

The Administrative Data Based Population, Household and Ethnicity Estimates (2016-2022) - Statistical Research publication found that using only available linked administrative data, a stated ethnicity could be found for 58.2 per cent of individuals, rising to 76.6 per cent when combined with Census 2011 data. Therefore it can currently provide substantially more population coverage for ethnicity than surveys but lower than what can be provided by a census. 

This research finds that both ABED methods estimates broadly align with Census 2022 results in representing the diversity of ethnic groups in Scotland, estimating similar distributions to Census 2022 results, including by age and local authority. Generally, the administrative data only method estimates are closer to Census 2022 results.  

Nationally, the differences between the ABED follow the same directional pattern for both methods - either consistently positive or consistently negative when compared to Census 2022 results – except for White – Scottish and Other Ethnic Group.

Both methods estimate a lower proportion in minority ethnic groups compared to Census 2022, with the exception of the Caribbean or Black ethnic group.  

For minority ethnic groups, that is groups identifying as Asian, African, Caribbean or Black or other ethnic group, the largest differences compared to Census 2022 are generally in the 20–64 age band. For White ethnic groups, the largest differences are in the 65 and over age band.  

While these findings are still at a research phase, it shows the potential for NRS to produce ethnic group statistics using administrative data. The ABED will continue to be developed, including through the inclusion of more administrative datasets to increase coverage.   

Project Aims  

National Records of Scotland (NRS) is responsible for Scotland’s Census, the data collection exercise used to prepare population, migration and characteristic statistics for Scotland. These data are of national strategic importance:  fundamental to resource allocation between UK countries, and across Scotland. Government, councils, the NHS, and public, private and third-sector users rely on NRS’ population statistics. 

Alongside other UK statistical organisations, (Office for National Statistics (ONS) and Northern Ireland Statistics and Research Agency (NISRA)) NRS are researching whether these critical population statistics can be produced using administrative data, alongside a census, with the aim of producing increasingly accurate, timely population statistics to best address user needs.  

This publication provides an update to NRS research on using administrative data and Scotland’s Census 2011 to produce estimates of the size of ethnic groups in Scotland. This work also aligns with one of the key lessons learnt from the 2022 Census to design use of administrative data from the outset for future censuses.  

UK Context  

The Office for National Statistics (ONS) is developing Administrative data based ethnicity statistics for England and Wales. Their latest publication, based on 2020 data, found that by combining ethnicity data from eight administrative data sources and the 2011 Census, they established an ethnicity for 84.9% of people in the 2020 administrative data based population base for England and 88.5% of people in the 2020 admin-based population base for Wales.  

The ONS has access to eight administrative data sources including England and Wales education data, HESA, NHS datasets and Birth notifications. Birth notifications in Scotland do not collect information on ethnic group.  

NRS work closely with the ONS and the Northern Ireland Statistics and Research Agency (NISRA) to share learning and ensure coherent statistics can be produced across the UK.   

Why Statistical Research Rather than Official Statistics  

For producers of official statistics, such as NRS, the term ‘Statistical Research’ is used to refer to research that is at an early stage of its development and would not meet the requirements for official statistics or official statistics in development. By using this term, NRS is able to formally publish material that can support further discussion and development. This publication presents the first iteration of our methodology and is still regarded as statistical research. As with our publications on administrative data research, NRS have provided a voluntary adoption statement to show how the principles of the Code of Practice for Statistics have been followed for this publication. 

These estimates should not be considered as a replacement for the Accredited Official Statistics Publication for Scotland.  

For official ethnic group estimates, please see Scotland’s Census 2022.   

Security and Confidentiality  

Security and confidentiality are of critical importance and we have taken action to ensure that personal information is safe. Only necessary data is collected for this project and when received it is stored securely, with access limited to authorised personnel. Personal information, such as names and addresses, are pseudonymised (so that original values cannot be determined) and datasets are brought together for linking and analysis after this has occurred. 

The administrative data based projects run by NRS have been scrutinised and approved by privacy panels and the National Statistician’s Data Ethics Committee. NRS is committed to transparency with users and the public in all aspects of this project.   

Table 1: Suppliers of the data used in the analysis and the datasets supplied.  

NRS would like to thank the suppliers in the table above as well as colleagues at the Scottish Government and eDRIS (part of Public Health Scotland (PHS)) for their ongoing support with this project. We would also like to thank all the stakeholders and peer groups, who have contributed their expertise and knowledge to support this work.  

All results are based on stated ethnicity, that is individuals with an ethnic group which is neither missing, refused nor not known. 58.2 per cent of individuals on the SIDD had a stated ethnicity, which increased to 76.6 per cent when supplementing with census 2011 data. 

The administrative data only method resolves the ethnic group for individuals by selecting from the source with the latest interaction date. If that is missing or not known, this method uses the next available. The administrative data and Census 2011 method introduces the Census 2011 record for individuals only if their resolved ethnicity from the administrative data is missing or not known. Figures 20 and 21 in the February 2025 ABED publication give a graphical overview of this methodology.   

If an individual has an ethnic group recording of refused in the ABED, this is retained and not replaced by an ethnicity from Census 2011. Census 2011 does not include children born, or migrants arriving in Scotland, after census date in 2011. Census 2011 provided an ethnic group for proportionally more individuals in rural areas where administrative data coverage was lowest.  

Small numbers may be suppressed in tables and charts where disclosure control applies. 

Tables, which include further detail than can be displayed in the charts, are available on the Scotland's Census website here.

National Level  

Both methods in the ABED produce estimates which broadly align with Census 2022 results. Generally, the administrative data only method estimates are closer to Census 2022 results. This may be due, in part, to the Census 2011 only containing individuals who were present on census day 2011.  

The largest ethnic groups in Scotland are White–Scottish and White–Other British as can be seen in Table 3. The ABED estimates a higher proportion in the White – British ethnic group under both methods. There is a White category available in administrative data that may affect the coverage of the remaining White ethnic groups, where individuals have the option to select this in administrative data but not in Census 2022.  

Table 3: Percentage of people on the 2022 ABED, and Scotland's Census 2022, for White-Scottish and White-Other British.  

For the minority ethnic groups represented in Figure 1, the ABED generally estimates a lower proportion than Census 2022, with the exception of the Caribbean or Black ethnic group.   

Proportionally, the largest differences between the ABED and Census 2022 are:   

For each of these groups, the percentage of people with a stated ethnicity is 1 per cent or less. 

These ethnic group categories are available on all source datasets, with the exception of White – Irish which is not available in HESA, which may affect the number of individuals in that group.  

Further investigation into the recording of these three categories is planned for future research, in particular variation across local authorities and age bands and between sources.   

Age Band  

*White – Other British contains White – English, White – Welsh, White – Northern Irish and White – British for the ABED. 

Other White Ethnic Group contains White – Irish, Gypsy/Traveller, White – Polish, Roma, Showman/Showwoman, White, and Other White Ethnic Group.   

Mixed or multiple ethnic groups contains Mixed – White and Black African, Mixed - White and Asian and Mixed – White and Black Caribbean for the ABED. 

Other ethnic group contains Arab, Arab Scottish or Arab British and Other ethnic group. 

Figure 2 shows the proportion of the ABED by Ethnic group and by age bands 0–19, 20–64, and 65 and over.  

Overall, variation in ethnic groups by age band in the ABED aligns with Census 2022, with larger proportions of people in White ethnic groups in the 65 and over age band.

For minority ethnic groups, the largest differences compared to Census 2022 are generally in the 20–64 age band. This is also the age range with the lowest coverage of ethnic group on the SIDD.  

For the 0–19 age band, the ABED estimates a higher proportion of people overall in White ethnic groups compared to Census 2022 and lower proportion in the minority ethnic groups.    

Local Authority   

The ABED broadly follows Census 2022 distributions at a local authority level, with estimates following the pattern of higher proportions of minority ethnic groups in urban areas.  

The largest percentage point differences between ABED and Census in the majority of ethnic groups are in the most urban local authorities: Glasgow City, City of Edinburgh, Aberdeen City, and Dundee City, due, at least in part, to these local authorities being more ethnically diverse than others. When taking into account population size, the proportional differences vary across all local authorities.  

The ethnic group which varies most across the local authorities is Black or Caribbean ethnic group as displayed in Figure 3. The ABED estimates a higher proportion in this group than Census 2022 in the most urban areas and a lower proportion in some, but not all, rural areas.   

Local authorities in Figure 3 are ordered from more urban at the top to more rural at the bottom.

Although the proportion of the population is small in the Caribbean or Black ethnic group, the graph shows where the estimates are lower or higher than the Census 2022 results.  

Some values in this graph are suppressed and not available due to the small reported numbers.  

The differences between the ABED and Census 2022 seen in Figure 3 could be for a few reasons, including the way individuals record their ethnic group on different sources or due to the coverage of stated ethnicity on the ABED.  

NRS will continue to research the differences between the ABED and Census 2022, including where individuals have different ethnic groups recorded on the ABED and Census 2022. Focus will be on urban areas where for some ethnic groups the differences between methods are more pronounced.  

The table below provides a description of the abbreviations used in this document.  

Public Health Scotland – Health Activity – Ethnic Groups  

Group A - White​  

Scottish​  

Other British​​  

Irish  

Gypsy/ Traveller​  

Polish​  

Other White ethnic group​  

Group B - Mixed or multiple ethnic groups​  

Any mixed or multiple ethnic groups​  

Group C - Asian, Asian Scottish, or Asian British​  

Pakistani, Pakistani Scottish or Pakistani British​  

Indian, Indian Scottish, or Indian British​  

Bangladeshi, Bangladeshi Scottish or Bangladeshi British​  

Chinese, Chinese Scottish or Chinese British​  

Other Asian, Asian Scottish, or Asian British​  

Group D - African​  

African, African Scottish or African British​  

Other African​  

Group E - Caribbean or Black​  

Caribbean, Caribbean Scottish or Caribbean British​  

Black, Black Scottish or Black British  

Other Caribbean or Black​  

Caribbean  

Other  

Group F - Other ethnic group​  

Arab  

Arab, Arab Scottish or Arab British​  

Other ethnic group​  

Group H - Not Known​  

Not Known​  

Group G - Refused/Not provided by patient​  

Refused/Not provided by patient​  

Statistical Research  

This publication presents statistical research and the methodology is still under development. We welcome any feedback from users on ways in which the methodology or data sources may be developed to improve the quality of these statistics in future years.  

Information on background and source data  

Further supporting documentation is published alongside this publication on the Scotland’s Census website.  

National Records of Scotland  

We, the National Records of Scotland, are a non-ministerial department of the devolved Scottish Administration. Our aim is to provide relevant and reliable information, analysis and advice that meets the needs of government, business and the people of Scotland. We do this as follows: 

Preserving the past – We look after Scotland’s national archives so that they are available for current and future generations, and we make available important information for family history. 

Recording the present – At our network of local offices, we register births, marriages, civil partnerships, deaths, divorces and adoptions in Scotland. 

Informing the future – We are responsible for the Census of Population in Scotland which we use, with other sources of information, to produce statistics on the population and households. 

You can get other detailed statistics that we have produced from the Statistics section of our website. Scottish Census statistics are available on the Scotland’s Census website. 

We also provide information about future publications on our website. If you would like us to tell you about future statistical publications, you can register your interest on the Scottish Government ScotStat website.  

Enquiries and suggestions  

Please get in touch if you need any further information, or have any suggestions for improvement. 

For enquiries, please contact [email protected] 

© Crown Copyright 

You may use or re-use this information (not including logos) free of charge in any format or medium, under the terms of the Open Government Licence. Further information is available within the Copyright and Disclaimer section of the National Records of Scotland website