Scottish Prisons Assessment and Review of Outcomes for Women (SPAROW): full report

3. Research aims and objectives

The research that is the subject of this report is set against the important background of the fundamental re-visioning of the women’s custodial estate and the core characteristics of women in prison in Scotland.

The primary objective of the SPAROW study is to evaluate the early impact and emerging outcomes of the new model of CCUs and the application of the SPS Strategy in the context of the CCUs. The research therefore aimed to provide an in-depth and focused assessment of the operation of the CCUs, the experiences of the women who live in them, the staff who work within them and the partners that deliver services into the CCUs. It also aimed to include the participation of SPS staff working in another prison establishment in Scotland and imprisoned women in that establishment who were not currently living in a CCU in order to gain understanding of their experiences of gender-sensitive and trauma-informed practice.

A set of pre-determined research questions were identified in the specification documents, although these were amended and added to by the research team during the research tender process. These additional questions drew on the existing research literature on women in prison and from our own experiences of working with this group. The additional questions were agreed by the research commissioners. A copy of the research questions can be found in Appendix 1 with the additions highlighted in blue.

The specification document for this work required that it comprise two phases. Phase 1 was to be an Interim Review which explores and captures emerging evidence on how the CCU model is operating in practice, and on how far gender-specific and trauma-informed principles and approaches have been implemented. It also specified that a Phase 2 on Emerging Outcomes should build on the findings from Phase 1, continuing to capture evidence, but also to capture women’s experience of living in a CCU, and whether and how their journey in, through and out of a CCU, might improve outcomes for them. This second Phase also explored outcomes experienced by SPS staff and delivery partners. Both phases utilise interviews with women, prison officers and managers and SPS delivery partners and identify examples of success and good practice as well as challenges and barriers. In the event, Phase 2 involved two stages of fieldwork – as we describe below.

Research design

In designing this research, we have taken a critical feminist perspective, incorporating the principles of ethical care, reflexivity, inclusivity, flexibility, activism, and empowerment. In practice, this means that the ethics of conducting the research are carefully considered and that as researchers we examine our own values, judgements and beliefs throughout the research process. It also means that we endeavour to create an inclusive and accessible research culture which shifts the balance of power from the researcher to those with lived experience and enables participants to shape the research process. This also means that the research design remains open and flexible to unexpected events such that it is able to accommodate new empirical materials which may benefit the research. This leads to research which embodies a commitment to transformative change in women’s lives. These principles, together with the nine SPS gender-sensitive and trauma-informed principles outlined above inform the overall methodological approach.

We adopted a Theory of Change (ToC) approach which entailed, first, drawing on interviews with senior SPS personnel involved in developing the vision for the CCUs and the use of key policy documents to set out the intentions and aspirations behind the vision. Second, we employed the full suite of research data to determine the extent to which the ToC is being realised in practice and the key barriers to implementation as planned.

Given the sensitivity around the evaluation and the high levels of media interest, a set of FAQs were developed by the research team. Following approval by the research commissioners, these were posted on the project web-page and can be found on the Scottish Centre for Crime & Justice Research website.

Research phases and fieldwork

Before commencing the research fieldwork, ‘getting to know you’ visits were undertaken in HMP Bella and HMP Lilias in July 2023. This afforded opportunities to meet the women living there and some of the SPS officers working there, as well as learning more about the layout and operation of the CCUs including the custodial environments and regimes, and links with external partners. Posters and flyers outlining the research and what participation would entail were displayed throughout the duration of the study in the communal Hub areas of the CCUs.

Following the ‘getting to know you’ visits, further visits to both CCUs were undertaken to discuss and develop interview themes and questions with women and SPS staff. This is congruent with our commitment to the use of a feminist methodology which accentuates women's experiences and the use of participatory methods in the design of research materials. These visits were intended to engage women and staff in a process of thinking about and sharing their experiences, and an opportunity to shape the research materials.

Two visits were undertaken to HMP & YOI Stirling, the new national facility for women built on the site of the old HMP & YOI Cornton Vale. The visits helped in gaining an understanding of the relationship between HMP & YOI Stirling and the CCUs and the physical layout of the new prison. The visits were also helpful in gaining insights into the processes for identification and selection of women from across the women’s estate who are considered suitable for transfer or progression to one of the CCUs.

As stated above, the specification for this study required that the research be structured in two separate but inter-linked phases. Phase 1 captures emerging evidence on how the CCU model is operating in practice, and on how far gender-specific and trauma-informed principles and approaches have been implemented. Phase 2 captures women’s experience of living in a CCU, and outcomes experienced by SPS staff and delivery partners. In practice, each phase provided data that allowed a longitudinal assessment of implementation and experiences.

Whilst we were required to structure the research in two phases, the fieldwork was conducted at three different time points. The budget was insufficient to allow continuous fieldwork, so a decision was taken to segment it into three stages. This would allow us to immerse ourselves into the workings of the CCUs for three periods, rather than just two and to strengthen the longitudinal lens of the study. The first period of in-CCU fieldwork, which focused on Phase 1, commenced in August 2023 and ran until November 2023. The second period of fieldwork was delayed somewhat by the HMIPS inspections of the CCUs that were carried out in January/February 2024; it thus commenced in March 2024 and ran until July 2024. The third period of fieldwork ran from September 2024 to February 2025. As indicated above, the latter two periods of research fieldwork comprised Phase 2 of the study. However, it should be noted that, outside the research fieldwork periods, members of the research team attended events in each CCU, such as displays and exhibitions of art produced by the women, coffee fundraising mornings, a family day, and International Women’s Day celebrations.

Most fieldwork visits to the CCUs were undertaken by two researchers working together, although on occasion three researchers were present. Researchers remained in the CCUs for between one and five hours each day during fieldwork periods. Some of the women living in the CCUs attended work placements during the week and so were frequently away when we visited. To ensure their inclusion, we often visited the CCUs at the weekend or stayed later in the evening. In total, across both CCUs, members of the research team visited the CCUs on a total of 42 occasions.

Ethics

Ethical considerations in this study are significant and determined the reflexive research design and implementation of the research. In addition to ensuring the informed consent of all participants, assuring anonymity/confidentiality, and the minimisation/avoidance of harm, other ethical issues related to the conduct of research in a custodial environment. These include the respective positions of the research participants (women in CCUs, prison staff and delivery partners) and likely barriers to their participation, safety and sensitivity to risk, and the power dynamics that exist between researchers and the researched.

All participants were provided with specific information about the research team, the aims of the research, why it was being done, and who the research commissioners were, in order to facilitate informed consent. Materials were also provided in Easy Read format. Participant recruitment required careful and ongoing consideration to avoid potential perceived pressure on participants. Researchers held frank discussions with participants about the research and their involvement, both before and after data collection and throughout the duration of the study. There was also inclusion of space in interviews for participants to discuss any questions or concerns they may have had about participating in the study.

We were very careful to acknowledge and uphold the autonomy, privacy and dignity of participants throughout the research process, with attention paid to the likely toll taken by interview participation, including debriefing and ‘warm-down’. Formal ethical approval for the research was obtained from the University of Glasgow College of Social Sciences Ethical Review Committee.

With the assistance of the University of Glasgow’s Research Regulation & Compliance Manager, it was determined that an Integrated Research Application (IRAS) for NHS Research & Innovation/Research & Development management approval to include the views of the NHS workers in the CCUs was not required. This was because participants would not be recruited via the NHS.

Methods of data collection

A multi-method approach focused on capturing the experiences and views of women, SPS staff and delivery partners was adopted. Each method is described briefly below: (see Appendix 2 for a more detailed description of the research methods)

• A review of SPS documentation (Phase 1) to provide background information on the aims and aspirations of the CCUs and the risk and progression processes for women.
• Fieldnotes (Phases 1 and 2) were taken within CCUs either during or following each visit. Rather than deploying formal overt observation we gathered systematic field notes but agreed with women and SPS staff in advance the kinds of features we might note.
• Online semi-structured qualitative interviews with SPS senior managers (Phase 1) were conducted to gain insight into the policy perspective on the CCUs. These interviews explored both the strategic and operational aims of the CCUs, interviewees’ reflections on the identification and selection of women for the CCUs and; the overall functioning of the CCUs. Four senior strategic managers (SSM) were interviewed; they are referred to in the report as SSM 1 - 4.
• Face-to-face semi-structured qualitative interviews with women living in CCUs (Phases 1 and 2) were undertaken to explore the key pre-determined research questions identified in the specification documents. These also allowed for a narrative approach to more fully elicit participants’ views and experiences and any issues that were especially important to them. A total of 33 women were interviewed. Throughout the report, women residents are referred to by the CCU in which they reside, an ‘R’ to denote that they are a resident, the fieldwork stage in which they were interviewed, and a number (e.g. BR2 005). Where women were interviewed in pairs, then they are distinguished as BR2 004:1 and BR2 004:2).

In addition, four larger group discussions were conducted with women (two in each CCU) where images from the photo elicitation exercises were used to stimulate reflection and a range of views. Numbers of women in attendance in these larger discussion groups ranged from five to eleven.

• Photo-elicitation with women in CCUs (Phase 2) was incorporated into group discussions as a means of evoking reflection and to enrich the interview experience. This method was chosen as a means of enabling women to articulate experiences that would, otherwise, have been difficult to articulate through talk-only interviews.
• Arts-based workshops with women in CCUs (Phase 2) were used as a participatory approach with groups of women living in CCUs in order to enable creative agency in articulating their unique experiences. Like photo-elicitation, this method was chosen as a means of facilitating reflection and dialogue, and also to access knowledge which is not easily expressed in words (Bagnoli, 2009; Barone & Eisner, 2012; Leavy, 2009).

  Three arts-based workshops took place in each CCU around 3-4 weeks apart in Phase 2. Women attended in groups ranging from six to eleven in size. Data and findings from Phase 1 were used to stimulate further dialogue on life in the CCUs, and to explore how things might have changed in the time between the fieldwork stages. Workshops involved an illustrator working directly with women to 'visually minute' what women were saying in words, metaphors and drawings that evoked their experiences, and then incorporated into large poster-style drawings. The worked-up versions were taken to the CCUs in a subsequent visit for the women to check accuracy and make any final changes (see Chapter 8 for the produced images).

• Face-to-face semi-structured qualitative interviews with SPS staff in CCUs (Phases 1 and 2) were undertaken to explore staff experiences and views about working within the CCUs. Interviews also explored how the CCU model might enable different relational, behavioural and cultural dynamics; successes/good practices and challenges/barriers to implementing gender-specific and trauma-informed principles/approaches to care and support for women in custody.
• A total of 40 SPS officers were interviewed. In the Report, they are referred to by the CCU in which they work (i.e. an L or B) an ‘S’ to denote they are staff, the fieldwork stage in which they were interviewed, and a number (e.g. LS2 002).

• Face-to-face semi-structured qualitative interviews with women in closed conditions in another prison establishment (Phase 2) were conducted in order to explore the key pre-determined research questions identified in the specification documents. The particular focus here was on women’s experiences of gender-specific and trauma-informed approaches in mainstream prison.

  A total of six women were interviewed. They are referred to using the same protocol as women in the CCUs, that is, GR 001, GR 003.

• Face–to-face semi-structured qualitative interviews with SPS staff working with women in closed conditions in another prison establishment (Phase 2). Interviews addressed how the CCU model is challenging working practice in other prison establishment(s) where women are detained, and in particular, how gender-specific and trauma-informed care and support of women is working in those other establishments. A total of four SPS officers from the same prison as the six women interviewed; three were operational officers and one was a Hall officer. They are referred to as GS 001 to GS 004.
• Online survey with SPS delivery partners (Phases 1 and 2). The views of delivery partners were sought with the aim of supporting the ongoing improvement of partnership working with delivery partners to design and deliver appropriate services that are gender-specific and trauma-informed. The first online survey was distributed to SPS delivery partners in November 2023 and generated 23 responses. The second was distributed in July/August 2024 and yielded just five responses. The findings assisted in identifying themes and issues to explore further in the study.
• Online focus groups and semi-structured qualitative interviews with SPS delivery partners (Phase 2). To elaborate on the survey results, all delivery partners who responded to the online survey were also invited to take part in an online focus group. This was to explore the factors that either facilitated/enhanced or limited their ability to engage directly with women, and to seek views on issues of community reintegration during sentence and support for transition at the end of sentences.

  Four focus groups took place with, respectively, two, four, five and six participants, and five 1:1 semi structured interviews took place with delivery partners.

Data analysis

Fieldnotes were taken throughout all stages of fieldwork. For the most part, these were descriptive notes recording sights, smells, sounds and impressions of the physical features and layout of the CCUs, and the ways in which the spaces were occupied and used by women and officers. Fieldnotes also included notes of interactions between women, between women and officers, and between officers. These included direct quotes, descriptions of interactions, and the researcher’s impressions of what these meant. Fieldnotes were useful in understanding participant meaning and provided rich contextual detail which aided in the wider analysis; they assisted in the identification of themes and facilitated preliminary coding of interviews and focus groups.

All interviews and focus groups were audio-recorded, with permission from participants (with the exception of one interview where the woman did not wish to be recorded, and so notes were taken instead). Following transcription, interviews and focus groups were coded using NVivo 12. The coding framework included emergent themes emanating from the data (including from the fieldnotes) and codes identified through the literature on women’s imprisonment and on gender-specific and trauma-informed practice.

Coding was undertaken by members of the research team to ensure consistency, sense-check ideas and explore different interpretations of the data. Reflexive thematic analysis (Braun & Clarke 2019) was undertaken. Data analysis was iterative, with researchers reviewing each other’s results and further coding/refinement for consistency. We attempted to ensure reflective and thoughtful engagement with the data and the analytic process whilst also acknowledging our own interpretations of patterns of meaning across the data.

Research fatigue and research hesitancy

As experienced prison researchers we were expecting some hesitancy from women and officers around engaging with the research and hence our materials and approach were prepared with this likelihood in mind. We set up the ‘getting to know you’ visits to introduce ourselves, to explain the research and what participation would involve and also as a means of circulating materials about the research. Participation in the research was entirely voluntary (and this was emphasised throughout) and our research materials were carefully constructed to be clear, transparent and informative.

Whilst we managed to interview all of the women residents in HMP Bella during the 1st stage of the research, the take up in HMP Lilias was lower. It emerged that the main reason for this was that women said they had already participated in SPAROW research and answered similar questions to what we were proposing. They were referring here to the SPAROW baseline study undertaken by the Scottish Government. They had heard nothing back from participating in that research and believed that ‘nothing had changed’ as a result of their participation. Whilst we did try to explain the differences between the two studies and that we were an independent team, it is clear that the proximity of both studies reduced women’s participation.

It would have been helpful to have received a list of women who underwent the SPAROW baseline study which, we learnt, concluded in July 2023. We started fieldwork shortly after in August 2023; it is therefore unsurprising that both studies were conflated in women’s minds. This is unfortunate and it would have been much more helpful to the evaluation if the baseline study had been concluded earlier (we understood at the inception meeting in May 2023 that the survey was closing imminently). Over time, as new residents arrived into the CCUs, this hesitancy became less pronounced.

We also encountered hesitancy from some officers in both CCUs. Initially we planned to hold focus groups with officers but, due to shift patterns, staff rotas and frequent high levels of staff absence due to ill health, the conduct of focus groups was rendered not feasible. In conducting this research, we tried to ensure built-in flexibility to take account of structural and organisational constraints in an evolving context. Whilst semi-structured interviews were suggested by several officers and agreed with us as a means of getting around the focus group practicalities, when it came to interview recruitment we encountered some resistance.

It transpired that this was due primarily to a general wariness about anonymity and confidentiality. It may be that the interview method is considered off-putting and exposing; whereas focus groups offer ‘safety in numbers’ and generate discussion between participants such that individual voices are less conspicuous. CCU officers are highly conscious of the media and public and political interest in the CCUs, which had already been subject to adverse media attention in some Scottish media outlets. As one officer maintained, the ‘high stakes’ media attention is a key reason for deterring officers to participate. Other officers spoke of ‘tensions’ and ‘opposing views’ within SPS about the viability of the CCUs and were wary of participating for fear of playing into these debates. In essence, this concerned officer perceptions of differing views about the need for CCUs and the new approach to women’s custody that they embody.

We always made arrangements in advance with managers to visit the CCUs to undertake interviews with officers and the women. However, on several occasions we were denied access by front desk operational staff who informed us that officers were no longer available for the pre-arranged interviews. This may be partly explained by low staffing levels as a result of staff calling in sick, but it did feel as if we were being dismissed without checking with more senior colleagues. To help address the situation, a Unit Manager who had participated in an interview sent an email to all officers saying the interview was enjoyable, that confidentiality was assured and encouraging their participation. This did result in proactive contact by a small number wishing to participate. However, there is a limit to such encouragement as participation must be voluntary. Participant recruitment requires careful consideration. Whilst every attempt was made to ensure free and informed consent, avoiding potential pressure (real or imagined) on women and officers to take part in the study, or to appease potential gatekeepers, we encountered some degree of hesitancy that featured throughout the study.

It is important to note that, even when officers did agree to participate in an interview, some were reticent in their answers to some questions, particularly those around risk assessment processes and the support they receive from SPS. Officers were visibly more relaxed when interviewed in staff offices rather than the Hub or one of the small meeting rooms. We suspect that this was because of the surveillance cameras in the public areas. In future, it will be imperative to always interview officers in staff offices where there are no surveillance cameras and privacy can be assured.

Limitations of the research

The limitations of the study mainly revolved around organisational and operational access. In addition to the hesitancy described above, a key limitation related to the constraints that operate within a prison setting – with locked doors and restricted areas. There were continuing debates over whether we, as researchers, were required to undertake the SPS Personal Protection Training (PPT). We were initially informed by SPS that we did not need this. However, at various points during the study we were asked to present confirmation of PPT completion on arrival at the CCUs. This impacted upon our access to and movements within the CCUs, as different line managers held different views about whether or not we required it. During the final (third) fieldwork phase we were told that we had to complete the training, but it was several weeks before we could gain training places for members of the research team.

Whilst we always called ahead to the CCUs to agree dates for fieldwork visits, often the women were not informed of this, and so were unable or unprepared to participate in interviews. Once in the CCUs, we were often restricted to the Hub (the communal area within the CCUs) which was not conducive to conducting interviews as there were usually other women, visitors and/or activities taking place there. If there were no women in the Hub at the times of our visits, we were entirely reliant upon officers to ask the women whether they wished to participate in interviews or activities with us. This often did not give us the opportunity to explain exactly what it was we were doing each time we were there. We attempted to get round this by providing flyers for the women explaining particular elements or research-based activities. As these were passed on to the women, this occasionally led to misunderstandings, such as the women being asked by officers if they wanted to be interviewed, when we were actually conducting arts-based workshops.

Sometimes we were allowed to sit in the Hub without supervision from officers and talk to any of the women who came in or could observe activities taking place. However, on a couple of occasions, due to the PPT issue, an officer had to remain whilst we spoke with women. This had an inhibiting effect. Where PPT was not raised as an issue, we were permitted to interview women one-to-one in an unoccupied side room, sometimes with an officer in the corridor. Women were always aware that they were under surveillance via CCTV cameras wherever they were in the CCU, and so this also tended to have an inhibiting effect.

Significant delays were experienced regarding which other prison establishments SPS would allow us to access to conduct interviews with SPS staff and women not working/residing in CCUs. Eventually we were given access to one closed establishment. The delays in being given clarity and access limited the number of interviews we were able to conduct, but nonetheless this was a valuable part of the research.

In addition to the delays outlined above, we were, for some considerable time, unable to obtain relevant data concerning women’s journeys in, through and out of the CCU for HMP Bella residents from SPS that was comparable to that received for HMP Lilias. Although we finally received this data in June 2025, it differs somewhat to that obtained from HMP Lilias and contains less detail. As such, we are unable to offer a detailed comparative analysis of women’s journeys in the two CCUs. Following the completion of fieldwork, we received data from Scottish Government Justice Analytical Services which provided a partial picture of the journeys of the women in HMP Bella.

There were two additional areas of exploration that may have provided useful insight and information into the women’s lives and the impact of the CCUs, namely following the women up in the community, and conducting research with their families. However, we believe both areas to be very sensitive, and to present significant ethical concerns. With women who have been released, seeking to conduct further research with them once back in the community presents a risk of continuing surveillance and disruption to their lives. Not all women have family visits, and for those who do not, this is experienced as difficult. We also know from other research that this can be a source of sadness and tension (e.g. Umeh 2025). Recruiting participants from those families who do visit will generate limited information about the role of the CCUs in enhancing family relations. Recruiting participants from families who do not visit presents a range of practical and ethical problems, even if women do give their permission to involve their families.