1. The Scottish Government is committed to tackling inequalities in Scotland. This has been a central element in its broad policy approach over recent years, as seen in the National Performance Framework, and underpinned by the Equality Act 2010.  In 2020, as a result of the Covid-19 pandemic - and concerns about its impact on different equality groups and the socio-economically disadvantaged - equality issues have come into sharp focus.
2. In order for the Scottish Government's ambition for the advancement of equality to be realised, it is vital that robust and comprehensive data on the characteristics of the people of Scotland (including equality and socio-economic disadvantage data) are collected. The availability and use of such data allows policies and services to be developed, delivered and adjusted in ways which align with the ambition to reduce inequalities.
3. In September 2020, the Scottish Government commissioned research to support its efforts to enhance the availability and quality of equality and socio-economic disadvantage data collected by public sector organisations in Scotland.
Aims of the research
4. The research had two aims. The first aim was to identify, describe and explore (i) what equality and socio-economic disadvantage data are collected by public sector organisations, (ii) how equality data are collected and stored, (iii) why or for what purpose equality data are collected and whether that purpose is fulfilled, and (iv) the barriers to equality data collection and to its use for the outlined purpose. The second aim was to use the findings of the research to offer insights into what works best in terms of collecting, utilising and safeguarding robust equality data and to highlight major barriers that could prevent collection or use of these data. The research also investigated views on whether the collection of equality data should be 'more extensively mandated under legislation' as has been suggested by some stakeholders in this field.
Research methods and approach
5. The research involved exploring the collection of equality data within a sample of Scottish public sector organisations and networks, gathering descriptive information about a range of identified data collections, and exploring the views and experiences of individuals working with selected data collections in a variety of capacities.
6. The research focused on organisations (and policy areas) with particular relevance to addressing 'structural inequality' and on 'public facing' organisations (i.e. service providers). The achieved sample comprised 27 organisations working across a range of policy areas (e.g. health and social care, justice, education and skills, etc.). In all, 55 datasets were included in the research.
7. The study involved: (i) a desk-based review of organisations' websites and publications, (ii) the collection of descriptive information about each selected data collection, (iii) qualitative interviews to explore issues related to the collection of equality data, and (iv) discussions with public sector equality networks.
What equality data are collected?
8. The research found that data on age and sex were collected in almost all of the data collections included in the study. Race and disability were collected in a large majority of cases. Religion and belief, and sexual orientation were collected in more than a third of cases. Marriage and civil partnership and gender reassignment were collected in around a quarter of cases, and pregnancy and maternity was collected in a small minority of cases.
9. One collection contained information about all nine (9) 'protected characteristics' and 14 further collections contained information about seven (7) or more characteristics. By contrast, seven (7) collections contained information about two (2) or fewer characteristics.
10. As far as socio-economic disadvantage data were concerned, three-quarters of the data collections contained postcode information (which can, at least in theory, be linked to the Scottish Index of Multiple Deprivation), around a quarter collected data on household incomes, and around a quarter collected data on other indicators of socio-economic disadvantage (e.g. property type, occupation, employment status, etc.).
11. In some cases, data on other equality-related characteristics were also collected. Care experience / care leaver and caring responsibilities were the most frequently reported additional characteristics collected; these were often collected in relation to education and other services with a primary focus on children and young people.
12. The question wording, accompanying definitions, and response categories were diverse, detailed and often complex. In general, there was a lack of standardisation in the way items are collected, in terms of: (i) the wording of / terminology used in the question, (ii) the definitions supplied to support the question, and (iii) the range of response categories offered. However, there were also a lot of similarities in the questions asked, and the differences were often detailed in nature.
13. Participants in the study often said that the Census 2011 questions had provided a reference point for developing the wording, definitions and response categories for the equality questions. There was also widespread use of a 'prefer not to say' response category (especially in relation to religion, sexual orientation, and gender reassignment).
Characteristics of equality data collections, and their purpose and uses
14. The (55) data collections explored in the research were highly diverse in terms of their content, operational context, and coverage. There was also variation in relation to the amount of 'infrastructure' (i.e. expertise, resources, and other supporting arrangements, both 'in-house' and external to the organisation) available to support their collection and use.
15. Around a third of the data collections were described as 'surveys' with the remainder described as 'administrative data'. Around a quarter were designated as 'National Statistics' (which means they meet the standards set out in the UK Statistics Authority Code of Practice). Approximately one-half were Scotland-wide, with a further one-fifth based in a single local authority. Two-thirds (including most of the administrative data collections) were described as 'ongoing' data collections.
16. The study found that Scottish public sector bodies collect and use equality data for a wide variety of purposes. In broad terms these purposes relate to: (i) fulfilling statutory requirements, (ii) developing, monitoring, evaluating and improving (national and local) policy, and (iii) developing, delivering, monitoring, evaluating and improving services.
17. The study found little evidence of data being collected, but not used. However, there was a range of cases where interviewees said the organisation could make more use - or better use - of the equality data they collected.
How are equality data collected and processed?
18. The methods used by public sector organisations to collect equality data included online, telephone, face-to-face, and paper-based methods - sometimes used in combination. There was also variation in (i) whether the information was provided to the organisation by the customer, or gathered and recorded by a 'third party' (i.e. a partner or external organisation, or individual(s) contracted to provide services), and (ii) the extent to which the collection of equality data was described as 'voluntary' or 'mandatory / required'.
19. Research participants also described the use of equality monitoring forms (EMFs). This took three main forms: firstly, EMFs formed the basis of the data collection; secondly, a separate EMF accompanied the 'main' data collection (note that in some cases these were returned separately); and thirdly, the EMF was integrated within the 'main' data collection.
20. The processing of equality data by public sector organisations (e.g. data input, quality control, data cleaning, data revisions and storage) is highly dependent on the infrastructure and resources which are available to support the particular data collection. Compliance with the Data Protection Act 2018 (DPA 2018) is the basic legal requirement, and the organisations participating in this study were very conscious of this.
21. The research heard about very varied processes. On the one hand, some data collections were gathered on paper (and in most, but not all cases, subsequently input to an electronic system). In other cases, there were very detailed arrangements for quality control and feedback (involving several iterations between respondents and the bodies collecting and reviewing the information), and collections were integrated into extensive IT systems and platforms, with complex and carefully developed data linkage arrangements in place.
Barriers and challenges in collecting equality data
22. The main barriers and challenges identified related to: (i) the personal and sensitive nature of equality data, (ii) practical, operational and / or methodological issues, (iii) difficulties relating to definitions and terminology, and (iv) shortcomings in the organisational culture, capacity and / or capability.
23. As far as the personal and sensitive nature of equality data is concerned, research participants explained that people can be reluctant to provide information about their equality characteristics, and may be concerned that this information might be used to discriminate against them. Members of staff can also feel uncomfortable asking people about their protected characteristics. There was also a view that individuals are less likely to disclose personal and sensitive information if they did not trust (i) the organisation asking for the information or (ii) the rationale given for collecting it. More generally, participants talked in terms of 'tensions' in relation to the collection of these data, the 'balances which needed to be struck' and the importance that the collection of equality data should be 'proportionate'.
24. In terms of practical, operational or methodological challenges, participants explained that the social and emotional context and /or the physical environment within which data collection takes place is not always conducive to the collection of high-quality data. For example, data collection can take place during a medical emergency or a police incident; outdoors in poor weather; in busy reception areas or other public areas; etc. Other challenges included: the lack of time available to collect data; the requirement to collect data by observation rather than by direct questioning; 'third party' data collection; and the multiple interpretations of some of the terminology used in equality data collection. This could be a particular issue for organisations providing 'frontline' services and dealing with the public in difficult or highly charged situations.
25. The third challenge - relating to definitions and terminology - was linked to a lack of standardised approaches in (i) developing questions, (ii) defining response categories, (iii) using terminology, and (iv) the definitions which underpin the questions. This was seen as raising issues of inconsistency across the public sector and between partner organisations. These challenges were raised most often in relation to data on sex and gender. Participants also said it could be difficult to keep abreast of changes in best practice in collecting equality data.
26. As far as shortcomings in organisational culture were concerned, a range of specific difficulties were described in which equality issues and the collection of equality data (i) were not prioritised or fully understood by senior management, (ii) were not embraced beyond the need to meet legal requirements, or (iii) had been side-lined by other organisational pressures.
27. In terms of the shortcomings in capacity and / or capability, some participants in this study highlighted (i) the lack of a 'joined up' approach to collecting and using equality data across the organisation, (ii) inadequate expertise for designing research tools, (iii) insufficient resources for data input, quality control and data processing, and (iv) a lack of analytical capacity to maximise the use / benefit of the data which were collected.
Enablers and facilitators in equality data collection
28. Research participants described factors that enabled or facilitated (or could enable or facilitate) the collection of high-quality equality data. These covered: (i) mainstreaming equality, (ii) the importance of being clear about the purpose of collecting data, (iii) the importance of training (for staff and third-party data handlers), (iv) using data collection methods that work, (v) building data management and analytical capacity, and (vi) improving guidance and developing networks for support.
29. Some participants highlighted their organisation's efforts to 'mainstream' equality. There was no standard approach to this, but suggestions for progressing this agenda focused on organisational structures, training, and leadership. Participants also emphasised the importance of taking steps to change organisational culture and behaviour. The requirement for public sector bodies to produce regular equality mainstreaming reports helped reinforce an emphasis on the development of equality outcomes.
30. In the main, organisations that were collecting high-quality equality data with good response rates had invested time and effort in developing clear communication with their customers. For example, they provided a short explanation, in plain English, about why they were asking for this information, what the information would be used for, and how the respondent's privacy in relation to the data would be protected.
31. Organisations that had high-quality equality data collections had also invested time in training staff and third-party data handlers to collect the data. This included the development of guidance, training materials, and 'scripts' that staff could refer to when speaking to customers.
32. In terms of data collection methods, organisations that collected equality data directly from their customers through an online system generally reported higher quality, more complete data. The advantages of online data collection included: (i) the facility for individuals to update their own information, (ii) the (in general) higher quality of self-reported data (compared with data collected through an intermediary or through observation), (iii) greater efficiency and cost-effectiveness, and (iv) the greater ability to build quality control procedures into data collection processes.
33. Where organisations had successfully built capacity for data management, analysis and reporting, this often involved the development of new IT systems and platforms incorporating better online data collection facilities; improved data cleaning; enhanced analytical capability; and more comprehensive reporting. These enhancements could create a 'virtuous circle' whereby the demonstration of potential benefits to the organisation of collecting and using the data could lead to improved collection and use.
34. Organisations with high-quality data collections - and those who wanted to improve their data collections - generally supported the idea of improved guidance and the development of networks. Suggestions about how this could be done included: (i) developing (better) standardisation and coordination of equality data collection - both within and across organisations, (ii) establishing and participating in (both internal and external) 'equality fora', and (iii) establishing a (nationally coordinated) 'go-to' place for guidance and support.
Views on making equality data collection mandatory
35. There was a range of views in relation to the question of whether equality data collection should be made mandatory. Whilst some participants were open to this suggestion and / or welcomed a move in this direction, others were more sceptical and had reservations and concerns of various types. Whether participants were in favour or not, they identified resource implications, if this option were to be pursued.
36. Those with positive or neutral views regarding mandatory equality data collection highlighted three main points. First, in highly developed administrative data systems there are detailed processes (including consultation with stakeholders) in place to determine what equality data are required. In such systems the quality of the data for mandatory fields tends to be very good. Second, in some specific cases (e.g. eligibility for financial or other support in relation to training or benefits) the data are already required (i.e. 'mandatory') for operational reasons. Finally, some organisations were keen to collect high-quality information about their 'customers' and so authorisation to collect this (if it were to be made mandatory by Scottish Government) would be welcomed / accommodated.
37. Participants who expressed concerns and reservations about mandatory equality data collection highlighted a range of issues including that: (i) an 'inflexible' or 'blanket' approach was not helpful; these kinds of decisions should be made locally by those who understood the environment within which data collection takes place and the requirements for the organisation to collect equality data, (ii) such a requirement would (or could) be disproportionate or inappropriate or the benefit of it for an organisation was not clear, and (iii) there could be a reluctance by organisations to collect equality data if previously undertaken analysis had revealed no policy impacts in relation to specific characteristics. It was also argued that any mandating of approach would present difficulties and / or would be inappropriate for services that deal with the public in more challenging situations.
Reflections on 'what works best'
38. Drawing on the research findings, four key themes were identified in relation to 'what works best'.
39. First, it was clear that the mainstreaming equality agenda - including the requirement for regular reporting - has been an important driver in promoting a focus on, and improving, equality data collection. Organisations had, for example, responded positively to this agenda by developing a supportive organisational culture and proactive leadership, investing in infrastructure, and ensuring a joined up, cohesive approach to equality data collection. In addition, it was thought that there was an important role for the Scottish Government to play in providing more guidance and improving the consistency of the data collected.
40. Second, there was compelling evidence of the importance of ensuring that all those involved in developing equality data systems were clear about why questions are asked, and how the information gathered will be used. This clarity is already a requirement of the DPA 2018. However, if equality data collection is to be improved, the need for clarity goes far beyond a (narrow) legal requirement. This clarity can build trust and improve disclosure. It is often prompted by - and developed as a consequence of - 'normalising' and 'mainstreaming' the equality agenda within organisations.
41. Third, this study found that public sector equality data collections were (not surprisingly) highly diverse in terms of subject matter. But they were also diverse in terms of their operational contexts. This variation was seen in terms of (i) the physical locations and emotional and social contexts in which data were collected, (ii) the individual(s) carrying out the data collection, and (iii) (partly as a consequence of (i) and (ii) above) the methodological approaches to collecting the data. This means that any approach to equality data collection must be carefully tailored to the situation at hand. The task of improving equality data collection cannot take a 'one size fits all' approach.
42. Finally, equality data collection is a highly complex undertaking. Larger organisations have a variety of internal and external networks and sources of advice they can access on an ongoing basis. By contrast, other organisations, including, but not exclusively, smaller organisations, felt there was a lack of a 'go-to' place for advice and expertise on equality data collection. This was notwithstanding that many of these organisations belonged to some kind of 'Equality Network' which they found helpful. Organisations with less equality data collection 'infrastructure' at their disposal had many questions about the 'nuts and bolts' of how best to undertake this work, and how to keep up with the (rapid) developments in this field.
There is a problem
Thanks for your feedback