Scottish Prisons Assessment and Review of Outcomes for Women (SPAROW): exploratory baseline study

Chapter Two: SPAROW Exploratory Baseline Study Methodology and Participants

Introduction

This Chapter discusses the methodology for the SPAROW exploratory baseline study, and women who participated.

Justice Analytical Services (JAS) Engagement with Women in Custody

The CCU model is founded on the principle that all aspects of the care of women in custody should be designed for women and take account of their likely experiences of trauma and adversities (SPS 2021). In line with this, JAS undertook small group discussions with women to discuss their thoughts on the most appropriate methods to engage and involve women in an evaluation of the CCUs^[3]. Women who participated provided important and thoughtful insights around surveys, creative methods, and group discussions, as follows:

• Surveys: women shared a preference for a survey method, and noted that surveys should provide space to explain answers more fully, and be dyslexia friendly (i.e. page tints and larger typefaces).
• Creative methods: women were also very positive about different creative ways to engage with an evaluation. Voice recording, journaling, drawing and photography were of particular interest.
• Group discussions: it became evident that women who were maybe more confident engaged more in the discussions, and women who were maybe less confident disengaged; which provided an important reminder to be aware of group dynamics (Bryman 2015; Clark et al 2021) during an evaluation, and ensure appropriate facilitation for all to participate.

As women said one of their preferred methods to participate would be a survey, for the baseline study JAS searched the available literature for potential validated surveys to adopt in the study. The Life in Custody (LiC) Study’s Prison Climate Questionnaire (PCQ) was found to be a methodological tool with potential to utilise in Scotland. The PCQ is specifically aimed at ensuring individuals in custody have a strong voice to share their experiences of the custodial care they receive^[4].

The Life in Custody Study and the Prison Climate Questionnaire (PCQ)

The Life in Custody Study, a collaboration between the Dutch Custodial Institutions Agency and Leiden University, is a large scale research project involving adult men and women in prison in the Netherlands. The aim of the study is to explore prison climate and quality of life in Dutch prisons through a validated survey instrument – the Prison Climate Questionnaire^[5] (Van Ginneken et al 2018; Bosma et al 2020). The LiC Study highlights that:

“A positive prison climate is expected to contribute to superior outcomes in terms of well-being, prisoner behaviour, treatment motivation and therapeutic change.” (Van Ginneken et al 2018)

The PCQ involves six domains of prison climate, and a number of measures/themes within each:

• Domain One: Relationships in Prison – staff/ prisoner relationships; prisoner relationships; procedural justice
• Domain Two: Safety and Order – safety and order
• Domain Three: Contact with the Outside World – satisfaction with visits during imprisonment; satisfaction with frequency of visits
• Domain Four: Facilities – sleep quality; quality of care; shop quality; complaints
• Domain Five: Meaningful Activities – satisfaction with activities; availability of meaningful activities; reintegration
• Domain Six: Autonomy - autonomy

The specific questions related to each of the measures/ themes are set-out in Chapter Four, alongside the report of the responses to these questions. Means scores in relation to overall domains and individual items are provided and compared pre- and post-move. Annex One provides information on the development of the Prison Climate Questionnaire, and detail on the six domains of prison climate.

The PCQ includes non-domain questions focused on: behaviours and victimisation; severity of imprisonment and future expectations; restorative justice; health and wellbeing; and, specific questions related to prison policy in the Netherlands. These questions were not used due to lack of relevance to this particular study.

Equality Characteristics

Equality characteristic questions were included to reflect equality legislation in Scotland on: age; disability/condition; transgender status; sexual orientation; ethnicity; and, religion. A comparison with women in the wider custodial estate at the time of the baseline study is provided, by average daily population (ADP) data, where data was available.

Qualitative Discussion

As noted above, women said that if a survey was undertaken they wanted space to explain their answers fully. Therefore, after each set of domain questions additional space was added to the PCQ to capture any further detail that women wished to share.

Expectation Questions

Finally, one section was added to the original PCQ survey in Scotland, to capture women’s expectations about moving to, and living in, a Community Custody Unit. Whilst living in an older prison establishment (pre-move survey, see Approach below) the questions focused on: feeling prepared to move to a CCU; what women were most looking forward to; and, what they thought they might find difficult. Whilst living in a CCU (post-move survey, see Approach below) the questions focused on whether living in a CCU was what women expected, what women thought was good and not so good about living in a CCU, and what could be improved. These questions are set-out in Chapter Three, alongside the report of the responses to these questions.

Approach to Undertaking the Exploratory Baseline Study in Scotland

The baseline study involved undertaking the Prison Climate Questionnaire and qualitative discussion with women in custody who would be moving to a CCU. The PCQ was administered as a paper based survey, printed in large arial typeface, with page tints available for women with dyslexia.

The baseline study was firstly undertaken in person, face-to-face, with women in custody before they moved to live in the Bella or Lilias CCU. At the time of the baseline study, women were accommodated^[6] in: His Majesty’s Prison (HMP) and (&) Young Offender Institution (YOI) Cornton Vale; HMP & YOI Polmont; HMP Edinburgh; HMP Grampian; and, HMP Greenock - hereon referred to as the participants’ “older establishment.” In this report the survey undertaken with women in their older establishment is referred to as the “pre-move survey”. The aim of the pre-move survey was to explore and understand women’s experiences whilst living in their older establishment (the baseline), before they moved to live in a CCU.

Secondly, about 6 months after women had moved to live in a CCU (or before they were liberated if less than 6 months from moving in), women who had participated in a pre-move survey were invited to repeat the survey again in person and face-to-face. In this report this is referred to as the “post-move survey”. The aim of the post-move survey was to explore and understand women’s early experiences whilst living in a CCU, to capture ahead of the commission of the main independent evaluation and inform that commission. The baseline study also enabled some early comparison of women’s experiences of living in an older establishment (the baseline) and a CCU.

Thirty-four women participated in a pre-move survey, with 23 of these women continuing to participate in a post-move survey – see Attrition limitation below. Annex Two provides more detail on how the study was undertaken.

Of the 34 women participating in the pre-move survey, the following numbers of women moved from these older establishments to a CCU: Polmont (16); Edinburgh (9); Cornton Vale (6); Grampian (2); and Greenock (1).

By way of comparison with Average Daily Prison (ADP) population data^[7], over the period of the baseline study^[8], 292.45 women were in custody, with over half accommodated in Polmont and Edinburgh, as follows:

Table One: Average daily population of women by establishment

HMP Establishment / Average Daily Population during baseline

Addiewell: <5

Barlinnie: <5

Bella: 7.3

Cornton Vale: 26.9

Edinburgh: 81.9

Grampian: 33.3

Greenock: 45.0

Lilias: 6.1

Low Moss: <5

Polmont: 89.3

Stirling: <5

Analysis and Reporting

All pre- and post-move survey responses^[9] were logged and recorded in excel. Data was then transferred, cleaned and sorted in R for analysis in excel. Analysis was undertaken by calculation of mean (average) scores which are presented in tables throughout the subsequent findings chapter. The pre- and post-move mean scores are presented side-by-side to highlight comparison of experiences of living in an older establishment and living in a CCU. In general, higher mean scores reflect a more positive experience of prison climate (Bosma et al 2020; Van Ginneken et al 2018)^[10].

As noted earlier, space was provided for participants to write down or discuss further detail on their experiences. The detail of the qualitative responses varied from a few written words, to the JAS researcher noting more lengthy discussion responses (where participants gave consent). The JAS researcher reflected with participants what they had consented to be documented on their behalf, to ensure they were content with what had been written. All the qualitative responses were typed into word, aligned with each domain theme/questions for analysis. This enabled the JAS researcher to identify themes and report thematically on what participants commonly conveyed, to add some depth to their quantitative responses. To interpret the qualitative data gathered, specific codes were identified and then grouped together where they shared an underlying meaning/pattern. This was very much an iterative process with codes and themes being refined as the analysis evolved (Bryman 2015; Clark et al 2021).

Limitations of the baseline study

Notification of Women Moving to, and Liberated from, a CCU

It was not always possible for the JAS researcher to be notified of women moving to a CCU, or of women who had been liberated from a CCU. Sometimes the decision-making process for a woman to move, and the arrangement of the associated logistics and preparation for their move, happened quicker than expected. For example, on some occasions women’s move date and/or time changed at short notice and it was not possible to arrange a pre-move survey before their move, or the JAS researcher was informed that a woman had moved to a CCU, after the move.

Where it was not possible to undertake the pre-move survey before women moved to a CCU, the survey was undertaken as soon as possible after a woman had moved to a CCU; of the 34 pre-move surveys, 16 were undertaken in a CCU. With the survey being undertaken in person, the researcher tried to minimise any bias in responses as much as possible by reminding women to focus on their experiences of their older establishment, and keep for example “their Polmont hat on.” However, it is acknowledged women may have been influenced by having moved to a CCU, and their early experiences of the CCUs therefore could have influenced how they responded about their experiences in their older establishment.

On a couple of occasions women accommodated in a CCU for a short period were liberated before it was possible to inform the JAS researcher, and therefore these women were unable to be invited to participate in a post-move survey.

Attrition

Related to the notification limitation above, attrition was also a limitation. As noted above, a total of 34 women participated in the pre-move survey, with 23 women continuing to participate and complete a post-move survey. There was more attrition from the Lilias CCU than the Bella CCU. Sixteen of the nineteen women who had moved to Bella continued to participate. Only seven of the fifteen women who had moved to Lilias continued to participate (see Table Two below).

Participation was voluntary and women could withdraw their participation if they wished to do so, without any need to explain. Sometimes on the day of the survey, women had changed their mind and/or did not feel well to undertake the survey. Another opportunity to continue to participate was offered, and if this was declined the researcher did not continue to contact these women, respecting their wishes.

It is acknowledged that this attrition could have shaped the findings discussed in subsequent sections of this report. It is not possible to be certain but, for example, women with more positive early experiences of living in a CCU continued to participate, whilst women whose early experiences were less positive chose not to continue.

Further, the attrition could also have been influenced by the survey method. The baseline study did not explore how well the survey worked for women in custody in Scotland, and potentially the survey did not work for women who withdrew.

Lack of Triangulation

As noted earlier, the baseline study did not involve SPS staff and delivery partners working in and with the CCUs, and therefore the findings cannot be triangulated with their perspectives. The main independent evaluation involved SPS staff, delivery partners, and fieldwork in another establishment, alongside women living in the CCUs, to explore in-depth through a variety of methods (noted above) how gender-specific and trauma-informed principles, values and approaches were being experienced and delivered in the CCUs; and how the Strategy for Women in Custody is working in practice in the CCU context (see Burman et al 2025). The independent SPAROW Evaluation report also draws on published research evidence to corroborate some of the evaluation findings.

Participants

The average age of participants in the baseline study was around 43 years. Most women reported they had a health condition or illness, identified as White Scottish, and were straight/heterosexual. Half of participants identified with a religion. Most women who moved to a CCU were serving short-term sentences (less than 4 years).

When comparing to available average daily prison population (ADP) data of women accommodated in the wider prison estate (at the time of baseline study), participants were slightly older on average than women in the wider population, who were aged on average between 30-44 years old. The ADP data also shows that most women accommodated in the wider estate identified as heterosexual and White Scottish, and were serving short-term sentences or on remand.

See Annex Three for further detail on participants’ equality characteristics, and comparisons with women accommodated in the wider estate at the time of the baseline study (where ADP data was available).