The Scottish Government

« Previous | Contents | Next »

Listen

CHAPTER TWO THE RESEARCH METHODS

INTRODUCTION

2.1 As outlined in Chapter 1, it was agreed by the Scottish Executive, in consultation with other key players, to pilot electronic monitoring as a condition of bail over a 2 year period starting in April 2005 in order to assess its feasibility in Scotland. Four pilot sites were identified for the use of electronic monitoring as a condition of bail, namely, the Sheriff Courts at Glasgow, Kilmarnock and Stirling and the High Court sitting at Glasgow. The evaluation covers the period from April 2005 until July 2006, the first 16 months of the pilots. This chapter describes the methods used in evaluating the pilots, including the collection of quantitative and qualitative data, an analysis of press coverage of electronic monitoring and bail and the economic component of the evaluation.

AIMS AND OBJECTIVES OF THE RESEARCH

2.2 The aims of the research were as follows:

• to evaluate the implementation of electronic monitoring as a condition of bail in replacing custodial remands and the potential impact on the prison population;
• to conduct a cost analysis of electronic monitoring as a condition of bail in contrast to those given custody/bail.

2.3 The key objectives of the research were as follows:

1. to identify how agencies prepared themselves across the pilot sites for this provision;
2. to monitor the extent and nature of training;
3. to assess levels of awareness of the provision among key groups;
4. to analyse the characteristics of all cases that meet the criteria for EM bail, regardless of whether it is applied for or granted;
5. to monitor both provision and remand generally;
6. to examine the effect that outside influences have on the ability of the accused to be effectively monitored, e.g., other court appearances and outstanding warrants;
7. to identify best practice in the use of electronic monitoring as a condition of bail across all agencies;
8. to monitor the breach rate and the action taken as a result of breach;
9. to analyse what behaviour constitutes breaches of electronic monitoring as a condition of bail, along with the individual characteristics of offenders;
10. to provide information on final sentencing decisions of cases involving electronic monitoring as a condition of bail;
11. to assess how the legislation is being used in each pilot site;
12. to establish how the legislation fits into the bail process and the implications for national roll out;
13. to analyse the cost of electronic monitoring as a condition of bail in comparison with custodial remand, factoring in any associated costs associated with breach of bail; and
14. to extrapolate from this analysis the costs to each agency of rolling out electronic monitoring as a condition of bail nationwide.

THE COLLECTION OF DATA

2.4 This evaluation employed a range of methods of data collection and analysis, both to describe the process of EM bail and to gauge outcomes. Qualitative interviews were conducted with sheriffs and judges, clerks of court, procurators fiscal, bail officers, the police, victims' agencies, defence agents, the electronic monitoring company, and bailees and their families. All interviews were tape recorded, with the exception of those conducted with judges. Quantitative data were collected, where possible on an ongoing basis, on the number and type of EM bail orders, previous and subsequent offending, demographic characteristics of EM bailees, and comparative data from 3 comparison sheriff courts, Greenock, Linlithgow and Edinburgh. Press coverage of the pilots was also examined through interviews with journalists and through scrutiny of the key national and local newspapers during the period of the fieldwork. Finally, a cost analysis of the pilots was undertaken relating to the economic implications of EM bail in the 4 pilot courts.

Qualitative data

2.5 The evaluation drew significantly on the views of stakeholders involved in the pilots, given their expertise and experiences to date of implementing EM bail. Thus, interviews were conducted with Steering Group members, with representatives of all the professionals involved and with bailees and their families.

Interviews with professionals

2.6 It was agreed to interview all 10 stakeholder members of the National Steering Group at an early point in the study in order to inform the research team's awareness of the practice issues emerging, any teething problems anticipated in the pilots, the logistics of the new legislation and training, personnel and resource issues. These members comprised 2 Scottish Executive personnel, a sheriff, 2 representatives of the Scottish Court Service, one defence agent, one social work manager, the manager of the electronic monitoring company (formerly Reliance, now Serco), a police superintendent and a procurator fiscal. The 3 criminal justice social work managers for Glasgow, Kilmarnock and Stirling were also interviewed. All these discussions were deemed useful to refine the research instruments, to negotiate the administering of information sheets and consent forms to (potential) bailees (see Annex 2) and to negotiate final access arrangements to key professionals and bailees and their families. Scoping interviews began in October 2005 and were completed in December 2005.

2.7 An additional meeting was held with Reliance management early on in the fieldwork period and with Serco management 3 months following the change over. Members of the research team also held meetings with the Scottish Criminal Record Office ( SCRO) and police personnel during the fieldwork period in order to finalise data collection methods for the quantitative data on bailees and the comparison sample.

2.8 As mentioned above, Steering Group members were asked for their help in identifying specific professionals for the research team to interview once the pilots had been operating for a year or more (i.e. as at April 2006). This small purposive sample cannot be deemed representative of all the agencies involved in the operation of EM bail in the 4 pilot sites nor of all the views of staff within each agency; nevertheless, they offer a likely range of evolving views and experiences of those staff and agencies directly involved with EM bail.

2.9 In addition to the scoping interviews, interviews were also held with the following professionals in the 3 geographical pilot sites of Glasgow, Kilmarnock and Stirling, and where appropriate, with specific personnel associated with the High Courts sitting at Glasgow and Edinburgh, all of whom had first-hand experience of dealing with EM bail applications, orders and failures to comply:

2.10 It was decided not to interview judges at the High Court sitting at Glasgow partly because of the turnover of visiting judges there, some of whom may have limited experience of EM bail according to court staff, and partly because the vast majority of EM bail applications emanate from the sheriff courts, where solemn cases are first heard. Although the number of defence agents interviewed was less than originally hoped for, it does include those practising in the High and Appeal Courts, cover the 4 pilot courts as well as the Appeal Court in Edinburgh, and all had experience of EM bail applications. Nevertheless, defence agents were not always amenable to being interviewed about the pilots because of industrial action over wider legal aid issues and other payments for court-related work at the time of the fieldwork.

2.11 Interviews with professionals tended to last between one and one and a half hours and included coverage of the following topics (see Annex 3):

Interviews with bailees and their families

2.12 A total of 91 bailees agreed - either directly by returning consent forms or indirectly by not responding to an opt-out letter - to participate in the research, which included agreeing to be interviewed about their experiences of EM bail. A further 11 actively declined to take part in the study. These individuals had been approached by bail officers preparing suitability reports or in retrospect by letter following a completed period of bail. An introductory letter and information sheet about the research, plus a consent form, were either posted to or handed to ex- or current bailees with the option of returning the signed form in a pre-paid envelope or giving this pre-paid envelope to the bail officer to forward to the researchers.

2.13 Although we had contact details for 91 bailees, it was only possible to interview 31 bailees or family members about their experiences of EM bail. This was because many no longer resided at the stated address or the phone numbers were no longer operational. A small number refused when contacted by phone or did not reply to messages. In particular, EM bailees who had been on EM bail prior to the interviewing commencing (in December 2005) proved difficult to trace in retrospect, not least given their transient lifestyles (often mobile phone numbers no longer existed; mobile phones had changed hands; or addressees had moved).

2.14 The breakdown of EM bailee and family respondents was as follows: 16 bailees (3 of whom were young women) and 15 household members ² (9 mothers, 3 fathers, one female partner, one sister and one cousin). Both bailee and a family member were interviewed in 8 cases, the bailee only in 8 cases and a family member only in 7 cases. Thus the respondents' comments relate to a total of 23 bailees. Respondents were located across the 3 pilot areas and related to Stirling bailees in 9 cases, Glasgow bailees in 9 cases and Kilmarnock bailees in 5 cases. The bailees had been on EM bail for between 2 weeks and 8 months, with 11 bailees both restricted from and to an address and consisted of previous and current EM bailees. The majority of bailee respondents had a range of previous offences and 13 of the total 16 bailees interviewed reported previous experience of custody. All respondents, both bailees and family members were positive about EM bail as an alternative to custodial remand.

2.15 Although initially the research team had hoped to undertake face to face interviews with the majority of bailees and their families in their own homes, gaining access to them proved difficult, and coupled with time constraints and safety issues, it was decided to continue interviewing bailees and family members by phone where possible. These interviews lasted on average half an hour and included questions relating to the following topics:

• previous experience of electronic monitoring and/or custodial remand;
• knowledge of EM bail and experience of the referral process;
• perceptions of EM company staff fitting and dismantling the tag;
• the conditions of the order;
• attitudes to breach and compliance;
• criminal justice and other commitments whilst on EM bail;
• the impact of the tag on lifestyle, relationships and offending behaviour; and
• overall perceptions of the appropriateness and effectiveness of EM bail.

The perceptions of victims

2.16 There is often an unspoken assumption that because EM bail ostensibly exerts more control over accused persons than other forms of bail, it should therefore offer a higher level, or an additional degree, of protection to victims. Within the literature on electronic monitoring, little is said about victims' views of tagging. In this research, given that the safety of known victims is an important consideration in the making of bail decisions, it was deemed important to gain some indication of what crime victims think about EM bail being imposed on someone accused of offending against them - even if such victims were not asked to have the technology fitted in their own homes as part of a 'restricted from' EM bail order. However, following discussions with the Scottish Executive, it was decided not to interview specific victims, but to gain their views through proxies, namely with a representative of each of the 2 Victim Information and Advice services based in the procurator fiscal offices in Stirling and Kilmarnock and with a representative of Assist, a voluntary organisation working in the pilot Domestic Violence Court in Glasgow Sheriff Court. It was the job of the VIAs and Assist to keep victims and witnesses informed of the progress and outcome of their cases, to support vulnerable witnesses in court and to refer them on, where necessary, to other organisations offering practical and emotional support.

Quantitative data

2.17 The following databases and files have been accessed to provide data on bailee characteristics, court processes and outcomes:

2.18 Court pro forma sheets that were sent to the Scottish Executive by each court were accessed by the research team, showing 271 applications for EM bail up until the end of July 2006; they provide some or all of the following information: court; COP number; verbal/written application; solemn/summary; refused immediately; date report requested/received; why refused (where given); whether granted/appealed and reasons. Not all clerks of court were able to complete pro formas in respect of their caseload relating to EM bail or to identify particular details on each case, and although the research team have supplemented the pro formas with additional records kept by court officials, Reliance/Serco and bail officers, it is possible that there may be other applications that have escaped recording on any of the agencies' databases.

2.19 Reliance/Serco have provided the research team with data relating to 63 closed cases ³ as at the end of July 2006 and include the following: SCRO number, PF number, Police number and Complaint number; breach information (number and types); date of order; timing of orders; estimated date of revocation; offence type; notifications to the police.

2.20 Data from court files (COP1 database) and social work records in Glasgow, Stirling and Kilmarnock have been obtained up until the end of July. These data supplement the court pro formas and Reliance/Serco database in providing the following information: COP number, date of custody court, date of EM bail hearing, outcome of hearing, sheriff seeking suitability report; sheriff reading suitability report; content of suitability report.

2.21 Access to police records using STORM was also granted by police statisticians, where available, up until the end of July 2006. This database provides information on the following: cases of failure to comply where this information was passed to the police and coded correctly; SCRO numbers; date of and reason for failure to comply; outcome of incident; length of time from notification of the failure to comply to final outcome; and number of police officers involved.

2.22 SCRO data provide information on the offending histories of those granted EM bail; and where possible, offences committed whilst on EM bail. SCRO data were also obtained post-July to match the final sample with a comparison sample from Linlithgow, Edinburgh and Greenock Sheriff Courts.

2.23 Using triangulation methods to build up a more accurate and fuller picture of the numbers being processed through the various stages of EM bail, it was possible to gain information on up to 306 applications for EM bail during the pilot period of April 2005 to July 2006. All quantitative data were entered into SPSS in order to undertake further analyses and calculate frequency data. Statistical tests (see Annex 4) were conducted to ascertain the similarities and differences between various groups within the overall sample population.

Limitations of the data

2.24 In terms of the above quantitative data, on average 14 per cent of these data are missing across all agency files (up to 8% missing data for complaint number, SCRO number, gender, age, presenting offence(s), date of application and suitability report outcome; and 40-60% missing data for identity of sheriffs at first and second hearings). Although the researchers attempted to reduce the amount of missing data by triangulation across all databases, this was not always successful. The Scottish Executive is aware of the limitations of quantitative data currently being collated by various criminal justice agencies, not least in relation to bail. The main issue for this evaluation was the use of complaint numbers, which apply to different cases (and within each case there may be more than one accused) rather than to different individual accused: in some cases applicants may present at court with more than one complaint file, and practice varies across clerks of court and across courts in relation to whether a separate pro forma is completed for each complaint number or for each 'case'. As each complaint number is not unique and is shared by each accused who is allegedly involved in the same offence, it has thus proved difficult to isolate only those who had applied for EM bail from those who had not.

2.25 However, the research team can be reasonably confident that cases where the sheriff asked for a suitability report to be carried out prior to the decision about bail being made are all included in this dataset, but where a sheriff immediately refused to consider EM bail, it is not always the case that a pro forma was completed. To supplement the information on the pro formas, the research team used the complaint number to access the COP1 database system in each pilot court to collect further information relating to the applicant (e.g., date of birth, SCRO number and presenting offences) as well as accessing the actual complaint files to find outcomes of trial, dates of applications for EM bail and sheriffs involved in both the initial application and second hearing for EM bail.

2.26 It has been highlighted in a recent report (Brown et al., 2004) that data on bail generally can be patchy and out of date. The report concludes that:

2.27 There have been major difficulties for the research team in accessing reliable, consistent and complete data on bail, custodial remand and breach statistics across the pilot and comparison courts; hence the need to 'triangulate' the data via several different sources, as mentioned above. Equally no data are necessarily held within the Scottish Executive's Justice Statistics Unit about the cancellation of bail orders or on charges proved in bail aggravated sentences. The above authors suggest that ISCJIS may be a way forward for centralising statistical and management information on the Criminal Justice system in Scotland in the future; however, that does not resolve the issue for the current study, where effective implementation, practice and evaluation are heavily reliant on quantitative data. The research team resorted therefore to manually retrieving much data from paper files to supplement the official data.

Comparison courts

2.28 Three comparison sheriff courts were identified, in consultation with the Scottish Executive, which matched the pilot sites in several ways: their populations and urban/rural breakdown were similar; their rates of custody overall and for various types of offence were similar, as were their lengths of custodial sentences. Table 2.1 below exemplifies these similarities with the comparison courts highlighted in bold. These figures are for 2004/05.

Table 2.1 Pilot and comparison court characteristics

Source: Scottish Executive (2006)

2.29 Glasgow Sheriff Court was matched with Edinburgh, Kilmarnock was matched with Greenock, and Stirling with Linlithgow. Kilmarnock and Greenock are both high incarceration cities, whilst Stirling and Linlithgow are both relatively 'rural' areas (although it was drawn to the Research Team's notice that Stirling has a higher than national average number of 'bench hours' per sheriff, and therefore 'rurality' does not necessarily equate with 'low usage'). The characteristics of the cohort of 191 accused in the comparison courts were analysed in terms of age, gender, number and type of previous convictions since 1989 and main presenting offence. All had been convicted of at least one offence in both solemn and summary proceedings between April 2005 and March 2006. The data were drawn from the Scottish Executive's court proceedings database.

2.30 Data matching of samples between the pilot and comparison courts was felt to be desirable for 2 specific reasons: one was to ascertain whether in fact EM bail was being used as a direct alternative to custodial remand (as opposed to being used to augment standard bail), and the second was to examine the various outcomes for both the pilot and comparison cases in terms of final outcomes and lengths of custodial remand/bail. The analysis of the comparison data is given in Chapter 5.

Observations

Attendance at policy/practitioner meetings

2.31 An additional element of the fieldwork included attendance by members of the research team at all Local Liaison Group meetings between November 2005 and September 2006. These meetings tended to be held every 3 months, allowing for observation and feedback about the research at a total of 4 meetings per pilot area. These meetings in particular proved invaluable in allowing for an observation of each group's deliberations on operational procedures and legislation as well as observation of inter-agency working. The meetings also gave the research team a greater understanding of distinctive local procedures and practices. Minutes and action taken as a result of these meetings were made available to the research team, as were minutes of Steering Group meetings, which were held on average every 4 months.

Attendance at custody court hearings

2.32 Most of the business of the courts in relation to custodial remand or bail occurs in the custody courts, which sit every day at a fixed time for those accused who had been remanded in police or prison custody the night (or last working day) before appearing at court. Research team members were able to sit in on 4 custody courts, 2 in Glasgow, and one in each of Kilmarnock and Stirling Sheriff Courts, to observe and to familiarise themselves with the process of applying for and/or being granted EM bail. One of these observations included a hearing of the Domestic Abuse Custody Court in Glasgow Sheriff Court, where the 2 resident sheriffs have used EM bail on a regular basis since the pilots began, not least to restrict an accused away from an address as well as to monitor movements of accused persons within their own homes.

Economic modelling

2.33 The aim of the economic component was to conduct a cost analysis of electronic monitoring as a condition of bail in contrast to those given custodial remand although the measurement of effectiveness is not straightforward in this field. The estimation of 3 related areas included:

• costs of EM bail in relation to data on the number of bailees, applications, appeals and breach;
• costs of custodial remand in relation to data on individuals held in custodial remand; and
• cost savings to the Scottish Executive and to the criminal justice agencies as a result of successful applications for EM bail.

Press coverage

2.34 The aims of the press analysis (see Annex 1) were to ascertain the characteristics of newspaper reporting on the EM bail pilots, to judge the extent to which it has presented a rational, rounded and responsible account of this particular initiative and to explain why the press coverage takes the form it does. This analysis covered the period March 2005 - November 2006, and required collation and analysis of press cuttings (including news items and/or editorials, augmented by Scottish Executive press releases) and interviews with journalists (2 on local newspapers in the pilot sites, 2 on national newspapers). Newspapers were initially analysed in terms of the following criteria:

• the specific prominence given to tagging in the context of the overall bail and custodial remand reforms;
• the use made by the press of official press releases;
• the use made - and the identity of - other key 'news and comment' sources;
• the use of human interest stories to bolster particular lines of argument;
• the use of photographs and logos.

CONCLUSIONS

2.35 The aims of the evaluation were to assess the implementation of electronic monitoring as a condition of bail in replacing custodial remands; and to conduct a cost analysis of electronic monitoring as a condition of bail in contrast to those given custodial remand.

2.36 Interviews were conducted with Steering Group members, professionals from the relevant partner agencies and with bailees and their families. Quantitative data was collated from various sources in both the pilot and comparison courts which provided a range of information on accused remanded in custody or bailed in areas outwith the pilot sites. This quantitative analysis explored the extent to which EM bail is being used as a direct alternative to custodial remand in the pilot courts and to explore differences in judicial processes and final sentences between the pilot and comparison groups. These data sources were not readily complementary or indeed complete, and given that different agencies had different methods of data collection and data protection, it proved difficult to build up a comprehensive picture of the whole process and outcomes of EM bail. To supplement the information on the pro formas (which were not always completed or were only partially completed), the research team also accessed databases pertaining to court records, police records and electronic monitoring company records, as well as paper files held by the courts and social work departments. Nevertheless, approximately 14 per cent of data were missing from agency files relating to EM bail. Overall, statistics on bail, remand, breach and final outcomes were also limited in scope and not readily accessible from various agency files. Numbers of applications which converted to actual EM bail orders were also relatively small, making meaningful comparison between and within the pilot sites difficult to achieve.

2.37 The cost element of the study enabled specific costs to be calculated for the various stages of the process of EM bail and thus enabled a comparison to be made between EM bail and custodial remand.

2.38 The press analysis of the first 21 months of the pilots not only explored the perceptions of journalists about electronic monitoring and bail but also informed how such perceptions could impact on the future success or otherwise of EM bail and EM more generally in Scotland.

« Previous | Contents | Next »