The Scottish Government

« Previous | Contents | Next »

Listen

An Evaluation of The Protection From Abuse (Scotland) Act 2001

CHAPTER TWO: METHODOLOGY

INTRODUCTION

2.1 The primary aim of this research was to monitor and evaluate the use and impact of the PFA Act, which came into force on 6 February 2002. It was anticipated that any evaluative study of this Act, coming so soon after its inauguration, would find minimal use of the legislation itself. Thus, it is more accurate to think of this study as a 'scoping' exercise with a more comprehensive evaluation of the Act best placed to occur once the Act itself and those people who both operationalise and seek help via its powers, have had more time to make use of the Act and its provisions. Nevertheless, the introduction of any new piece of legislation requires early investigation of its use and potential effectiveness in order to ascertain the extent to which further work requires to be done with regard to, for example, enhancing the profile of the Act, its powers and its provisions.

DESIGN AND METHODOLOGY

2.2 The overall design of this study was comparative and comprised of 2 distinct elements. Firstly, in order to ascertain the extent to which the Act was being used, it was decided that the number of applications for civil protection orders brought under the Act would be examined across 4 sheriff courts in Scotland. The courts selected by the Scottish Executive, in conjunction with the research team, for inclusion within the study were Glasgow, Stonehaven, Dumbarton and Stirling. This sample was considered as offering a balanced combination of demographic factors. Secondly, in order to gauge use of the new Act, it was critical to compare it with applications for orders for protection before the Act came into force. Thus a comparative examination of applications for protection from abuse brought prior to the 2001 Act under existing provisions and of those applications made after the 2001 Act came into force was undertaken.

2.3 The research team also recognised that whilst an examination of official sheriff court records would provide important data in relation to the use of the new Act, other sources of data would provide invaluable information triangulating the court documentation. Thus in order to fully evaluate the new Act it was important (a) to ascertain the views of those operationalising the new Act and by implication the old acts (b) to ascertain the views of those using the new Act and by implication the old Acts and (c) to investigate any criminal proceedings taking place should the new Act be breached, a critical element of an examination of the impact and effectiveness of the new legislation.

2.4 Thus, the overall design of the study was comprised of 4 different and complementary elements thereby allowing the 'triangulation' (Denzin, 1970) of both quantitative and qualitative data, which included:

• interrogation of official civil court records processed in both pre and post Act time periods across 4 identified courts
• scrutiny of breached cases processed through criminal courts
• examination of the views and experiences of court professionals operationalising the Act, including police, sheriffs, procurators fiscal and solicitors
• examination of the views and experiences of those most likely to seek protection under the Act

2.5 From the above design, a number of specific objectives for the study were outlined:

• to identify the number of applications for interdict (with or without a power of arrest) or non-harassment orders as a result of domestic abuse processed though the 4 identified courts in the 3 month period prior to the implementation of the new legislation (Nov 2001-Jan 2002)
• to identify the number of applications for interdict (with or without power of arrest) and non-harassment orders as a result of domestic abuse processed though the same 4 courts in the 3 month period after the implementation of the new legislation (Feb-April 2002). However, the Act of Sederunt (Ordinary Cause Rules) Amendment (Applications under the Protection from Abuse (Scotland 2001) Act) 2002 ⁹ did not come force until 8 March 2002, therefore no PFA orders could be sought in February 2002. Given this, the research team collected data for an additional month (May 2002), thereby ensuring that more applications made under the 2001 Act were included within the data-set
• to establish a series of baseline measures in order to examine the cases processed through the courts before and after the implementation date
• to identify dominant patterns both generally but also in relation to each of the 4 courts with specific reference to identified baseline measures
• to ascertain the views of women on their experiences of using the new legislation, particularly they extent to which they feel the law has 'protected' them
• to ascertain the views of key court personnel on their experiences of implementing the new legislation in terms of its effectiveness and the problems encountered in its application
• to examine a sample of PFA Act cases which had been breached and processed through the criminal courts

Using documentary sources in legal research

2.6 A major part of the overall design of the study was the interrogation of court documentation. It is therefore important to say something about the use of documentary sources in social and legal research before discussing the methodological approach taken in relation to the civil court records themselves.

2.7 The use of documentary evidence to advance knowledge about the social world is growing (Lee 2000; Scott 1990; Evans 1997; Featherstone 2000). With regard to legal research, documents can provide one of the primary sources of evidence and their use, particularly in conjunction with interviews with key informants, can provide invaluable data. However it is unhelpful to view documents as containing the 'truth'. Rather than simply being read as straightforward accounts of a phenomenon, documents and the evidence contained within them rely on the assumptions, conceptions, definitions and priorities of those who 'produce' them. Thus any critical analysis of documentary evidence should be mindful of this important consideration. Legal documents are circumscribed by legal discourse but in evaluating any piece of legislation it is problematic to scrutinise such legislation solely through the 'lens' of the law. In order to develop more effective legal and policy responses, the examination of legal processes to social problems requires systematic understanding and investigation of the many factors which impact on the causes, effects and consequences of the phenomenon and the individuals primarily affected by it.

Initial Writ

2.8 This is of relevance when analysing data gathered from documents contained within court records. The research team in this study gathered data on civil applications from the initial writs contained within court processes. An initial writ contains three sections. The first section sets out what the applicant or pursuer wishes the court to grant, referred to as 'craves'. The second section contains information in support of these requests and the third section sets out the legal basis upon which the requests should be granted. On the basis of the information provided by a client, a solicitor drafts an initial writ. This involves a process of translation whereby the client's wishes and story are translated into legal craves with relevant supporting evidence. This process involves the solicitor advising the client as to which type of legal order may be best able to achieve the client's aims. The advice given will, therefore, depend both on the knowledge of the solicitor of the options available and also the information given by the client.

2.9 The information contained within an initial writ to support the craves might not include all of the information provided by the client to the solicitor. The sheer volume of information available and a preference for the best evidence, may necessitate that some information is excluded. The best evidence would be that which best supports the granting of the order(s) craved and, could be proved in court should the other party decide to defend the action. As a result, the initial writ must be viewed as the product of a process whereby the information provided by a client is presented within legal norms. The woman's relationship with the man and her experience of domestic violence and abuse is translated into an account which can be proven in court and which meets the legal requirements of the civil protection order which is being sought. The initial writ should not be viewed as a comprehensive record, which reveals, in the context of this study, the full history of the parties' relationship and the violence and abuse suffered by the woman. Instead the information within the writ must be viewed as a partial account of the parties' relationship and the violence and abuse suffered by the woman that has been drafted for the particular purpose of seeking a civil protection order.

Access to and collection of data from sheriff court civil records

2.10 Prior to the research being commissioned, the Scottish Courts Service requested that sheriff clerks dealing with civil business in all courts in Scotland identify and record cases which would be relevant for the study. This process began in October/November 2001 prior to the research commencing. Initially sheriff clerks were recording only the first crave in any initial writ: the researchers identified that a request for an interdict or non-harassment order may often appear as an ancillary crave and as a consequence all ancillary craves were also recorded thereafter. This was of particular relevance where an interdict to protect against abuse was sought as an ancillary crave in an initial writ for divorce.

2.11 Lists of civil court actions where the writ craved for an interdict, interdict with power of arrest or NHO, were supplied to the researchers by Scottish Courts Service. Technical difficulties led to delays in obtaining this initial information and subsequent delays in the collection of data. When lists were obtained, the researchers forwarded these to the sheriff clerks in the relevant courts and attended on agreed dates to collate the data. This involved between 2 and 4 visits to each court.

2.12 In order to address the aims of the research, both qualitative and quantitative data were collected from the court records. Each court record included an initial writ which sets out the form of protection the applicant was seeking and provides a statement of prior conduct by the other party to support the desired order sought. Some court records contain other relevant information e.g. affidavits of those who have witnessed abuse. Importantly, information contained within writs is limited often to information required for court processes. Other data may be reported in individual writs but is not systematically recorded. We were interested in capturing data relating to many aspects of women's and men's experiences. Thus as much data as was available was gathered from the court records including: biographical details of both pursuer and defender; details of the application including the number and type of craves; details of the evidence upon which the craves were based (i.e. the incidents of violence and abuse); and details of the outcomes of the applications. Whilst the quantitative data permitted the 2001 Act and the use of previous provisions to be enumerated, a comprehensive picture of the abuse experienced by the applicant can best be captured by the gathering of qualitative data. Thus quantitative data was supplemented by qualitative data taken from the court records: this was invaluable in providing an insight into aspects of individual women's experiences of violence and abuse and responses to it.

2.13 Data from a total of 123 court records was gathered across the 4 sheriff courts. Using SPSS quantitative software package, a codebook was constructed and used by the researchers to record data directly into a laptop computer. Qualitative data taken from the case files was recorded in a Word file.

Access to and collection of data from sheriff court criminal records

2.14 The research team were of the view that a thorough evaluation of the 2001 Act required data to be collected on civil protection orders that were breached. This would provide information on the extent to which civil protection orders provide effective sanctions in relation to the behaviours of perpetrators they seek to influence.

2.15 We also wished to identify how breached orders were responded to. The importance of monitoring civil protection orders and the breach of these orders had not been anticipated in advance of the tendering process and therefore, the Scottish Courts Service had not been asked to instruct sheriff clerks to 'flag up' criminal cases. Due to data protection restrictions and the short time scale for the collection of data, it was not possible to monitor the civil orders, which had been identified by the sheriff clerks to ascertain if these were subsequently breached. A period of prolonged access negotiations commenced which resulted in the initial period of research being extended. Both the Crown Office and Scottish Courts Service informed us that it would be impossible to identify or 'flag up' breached interdicts from the records of criminal prosecutions. This is because the prosecution of a breached interdict will proceed on the basis of the crime committed e.g. breach of the peace rather than e.g., breach of interdict, as the latter is not a crime. It is impossible to differentiate a breach of the peace charge arising from an incident where a civil protection order is breached from any other instance of breach of the peace. The only breached civil protection order which can be easily identified is a breached NHO. Breaching an NHO is itself a crime under section 9 of the PH Act 1997 and is charged as such.

2.16 Other methods of identifying relevant criminal cases e.g., 'flagging' by sheriff clerks or procurator fiscal deputes were suggested by the researchers but it was not possible to progress these. In initial discussions to secure access to criminal cases Crown Office personnel asked for confirmation that Crown Office obligations under the Data Protection Act would be met by the research. Until this matter was resolved by reassurances of anonymity etc. the question of identification of relevant cases was not addressed. Identification of relevant cases proved to be rather complex and resulted in a period in excess of 12 months passing, before the matter was resolved. During this time the research team was in regular contact with the staff at the Crown Office and the Executive who were endeavouring to secure access to this very important data.

2.17 The problems of accessing criminal court records were partially resolved following the continued efforts of the research team to secure access to some data informing this critical element of the study. Eventually Strathclyde Police confirmed that they could identify from their own records, cases where a civil protection order was breached and a report was sent by the police to the procurator fiscal. Thirty breaches were identified as relevant by the police and sent to the Crown Office. These breaches had occurred between October 2001 and September 2002. To avoid additional delays that would have been occasioned by Data Protection Act requirements if the researchers were given direct access to the papers, a compromise was reached whereby the research team designed an information template for the collation of data from the cases where a civil order had been breached. This was completed by Crown Office staff on behalf of the researchers. Whilst this type of data collection is not ideal, the researchers having no control over data recorded, it was deemed to be the best possible compromise. The research team liased with the staff member who was collecting data at Crown Office to ensure that the data collected met the requirements of the research, as far as was possible.

2.18 Strathclyde Police identified 23 relevant cases. The Crown Office supplied data on 18 of these 23 cases, 5 being omitted as the proceedings were ongoing or due to the possibility of identifying the parties. On receipt of the 18 completed templates, it became clear that the police had identified 15 breached non-harassment orders (2 of which related to neighbourhood disputes and were irrelevant to this study), one breached matrimonial interdict and 2 breached interdicts which were granted under the Protection from Abuse (Scotland) Act 2001. Much valuable data was collected from the templates detailing the breached protection orders and interviews with procurators fiscal. However, a more comprehensive study of work in this area is required before firm conclusions can be drawn concerning the effectiveness of civil protection orders.

2.19 One of the original research aims was to identify rates of reported abuse and the numbers of cases in which powers of arrest were exercised by the police. Scrutiny of the court records revealed that this information is not systematically gathered. However, information on the effectiveness of civil protection orders and how often they were breached was explored in interviews with police officers and procurators fiscal (discussed later), thus some impressionistic data was gathered here.

2.20 The access difficulties encountered in this study, particularly in relation to the criminal cases, resulted in the duration of the project being extended by 11 months. Adopting a system whereby sheriff clerks identify relevant cases in advance of, and concurrently with, any research study in progress should reduce access difficulties in future. This would also allow researchers direct access to court records. Such an approach would, however, result in cases marked 'no proceedings' being omitted from any study and may exclude the investigation and analysis of any difficulties encountered in bringing prosecutions.

Postal questionnaires to women using the legislation

2.21 Information from individuals who had experienced domestic abuse and sought protection either from the police or by using civil protection orders was essential to supplement the data collected from the civil court records. The research team were of the view that gathering information from individuals who had experience of using the legislation would provide a source of rich interpretative data. Access to both women and men who might have used the act was planned. The terms of the Data Protection Act (1998) prevent information in court processes being used to contact individuals. To overcome the legal restrictions of data protection, we used existing contacts in men's and women's organisations and family law solicitors to establish contact with potential users of the legislation. We reasoned that asking individuals to talk about their experience of the court process and violence in an interview may be distressing, therefore, we chose a postal questionnaire to collect data from people, a method which maximised anonymity and which might be potentially less painful for participants (Kelly et al 1994; Finch 1984). Access was restricted to individuals resident in Glasgow, this city being chosen (a) because of the large population and (b) because the research team had contacts within relevant organisations.

2.22 A postal questionnaire was constructed and piloted with the assistance of women's organisations. Women victims of domestic abuse living in heterosexual or same sex relationships can use the services of Women's Aid. No specific organisations provide services to male victims of abuse, whether in hetero or same sex relationships, however, such men may seek help from organisations providing services in a health care setting. We therefore contacted one such organisation who agreed to participate. Questionnaires were sent to all organisations who had agreed to distribute them to potential respondents. These organisations included, 6 women's organisations, one men's organisation and 6 solicitor's firms. Ninety-five questionnaires were sent to the aforementioned organisations for distribution and a total of 32 (34%) were returned. All of these were completed by women and no responses were returned by men. ¹⁰ Women returning the completed questionnaire received a 10 fee.

2.23 Both quantitative and qualitative data was gathered from the questionnaires which covered a number of areas of women's lives including: biographical information; details of the violence and abuse; experiences of agency responses including police and courts; knowledge of the new PFA act and experience if used. Quantitative data was analysed using SPSS software computer package and qualitative data recorded from the questionnaires was recorded on a Word file.

Qualitative interviews with key court and police personnel

2.24 In order to augment data gathered from court records, a total of 18 interviews with key personnel involved in using the existing provisions and those introduced by the 2001 Act were conducted. Professionals had the opportunity to comment on the pre-existing and post PFA act provisions. Interviews were conducted with a total of 18 individuals: 4 sheriffs, 4 procurators fiscal ¹¹ and 4 police representatives (including uniformed officers and domestic liaison officers) from across the 4 sheriff court areas. Six solicitors, from 3 of the sheriff court areas, ¹² who were engaged in a range of practice settings were also interviewed. Interview schedules designed to gather purely qualitative data were constructed. A number of general and particular themes (relating to the roles and tasks of the different professional) to be explored in interviews with all professionals were identified, including: knowledge and experience of the new Act and its provisions, cases suitable for the Act, general experience of working with domestic violence cases, experience of obtaining legal aid, breach, evidence, interim orders, operating powers of arrest and the need for further legal or non-statutory reform. The 4 procurators fiscal depute were interviewed specifically in respect of decisions to prosecute breaches and the compulsory appearance of a perpetrator before a sheriff where a decision has been made not to prosecute. Interviews were tape-recorded and transcribed. Qualitative data was analysed manually.

« Previous | Contents | Next »