Scottish Government

The Role Of Mediation In Tackling Neighbour Disputes And Anti-Social Behaviour

4. RESEARCH METHODOLOGY

Data collection and analysis

4.1 In autumn 2002, 100 neighbour dispute cases from four mediation service providers across Scotland were accessed. Fifty cases were obtained from two community mediation services. The geographical catchment area of Mediation Service A was a city and Mediation Service B covered a large mixed urban/rural area. Fifty cases were accessed from in-house mediation services of two local authorities, Mediation Service C and Mediation Service D, both of which had mixed urban/rural catchment areas.

4.2 Community and in-house mediation services were asked to keep additional records of time spend on all neighbour dispute cases during a specified period (generally one month). In some cases, a further period was required in order to capture a sufficient number of cases where mediation activity had taken place. The researchers examined cases files and took details of the nature of the dispute. Interviews were then held with staff to obtain data to allow the research team to calculate hourly rates and costs per case. In the local authority services, case costs were calculated by the service, due to the complexity of mediator costs where different mediators had different hourly rates. For some organisations, it was possible to compare this with costs per case calculated from total budget and caseload. Mediation services obtained permission from clients for their records to be used in the research. Where possible, outcomes were checked using services' own follow up procedures for cases where agreements are reached (some services have a fixed follow-up period, usually three months, others set a variable review period with the parties).

4.3 For legal interventions, meetings and telephone discussions were held with several local authorities, which, according to Chartered Institute of Housing survey data, appeared to take the highest numbers of legal actions. Two local authority housing investigation teams agreed to allow researchers to examine their cases. These data were then supplemented by examination of legal files and interviews with solicitors. A further two such teams submitted anonymised case records to the research team. Four local authorities without specialist investigation teams were also involved. In these areas, the researchers examined legal section files and interviewed housing officers and solicitors to gather information, and in some cases housing officers and solicitors also provided their own calculations. Several local authorities were able to provide information on only one or two cases. A total of fifty cases were accessed across eight geographical areas, including one city, two predominantly rural areas, and five mixed urban/semi-rural areas.

4.4 As for police records, in one area, it was possible to prepare summaries of ASBO cases for senior police officers who calculated time spent on each case from their records. In a second area, police officers were able to give data relating to a number of anonymised cases, but these could not be linked directly to housing services' data. Environmental health officers in one local authority prepared summaries of cases detailing their involvement. Contacts with other environmental health services and police forces did not result in access to additional data. A survey was conducted of 41 housing associations; from nine responses, three reported having used community mediation, but only one association reported having taken any legal action for anti-social behaviour in recent years (through a local authority investigation team which took part in the present study).

4.5 Analysis was carried out using Excel and SPSS to determine costs per case and to make comparisons between types of case. A series of assumptions were made in order to arrive at costs per case, and these varied between the services contacted depending on the nature of their data. These are shown along with the case information in the appendices. The costs given do not include overheads such as buildings costs, central administrative services (payroll etc.), employers' costs (NI and pensions). The costs given are salary (including overtime) and travel costs that can be attributed to particular cases.

4.6 Interviews were held with people who had taken part in mediation. Their names were given by mediation services who had written to ask for their agreement to take part in the research. Seventy participants from three mediation services agreed that their contact details could be provided to the researchers. Twenty-seven telephone interviews were held with mediation participants, around two-thirds from local authority services and one-third from a community mediation service. One party from each case was interviewed.

4.7 Twelve housing officers, from four local authorities, who make referrals to mediation were also interviewed as to their views on reasons for refusal of mediation, and also to find out whether they might be able to contact 'refusers' on our behalf to take part in the research (none were able to do this). Two mediation services provided contact details for 11 potential interviewees who were parties to neighbour disputes and had decided not to take part in mediation. Of these, five were interviewed.

Quality of the data

4.8 Previous studies have highlighted the methodological challenges created by the definitional and conceptual issues noted above, and the general lack of data among the organisations involved. The problematic nature of cost benefit analysis of mediation has been reported upon by previous researchers (Dignan, Sorsby and Hibbert 1996). In addition, confidentiality and data protection considerations present difficulties for research such as this, which aims to examine specific dispute cases. For mediation services, confidentiality is central to their practice.

4.9 For data protection reasons, data from the police, with the exception of a small number of data from one force, was provided in a way which made it impossible to link it to particular cases.

4.10 The researchers obtained access to cases from mediation services with relative ease. Particularly for in-house services, which lack the standardised recording systems of community mediation services, there was a lack of harmonized data collection on presenting issues for cases and a range of criteria and codes used for outcomes. Where the quality of data needed to be enhanced, mediation service providers were asked for explicit clarification to ensure that, as far as could be possible, the researchers could compare like with like, between mediation cases. Mediation services keep limited information about the nature of disputes on file; this was a barrier to accurate categorisation of cases. Some, though not all, mediation services, for confidentiality reasons, provided names of interviewees in such a way that data about outcomes from interviews could not be linked to recorded outcomes, although valuable participants' views could nevertheless be gathered.

4.11 There is little systematic recording of outcomes of legal action. One of the specialist investigation teams involved in this study keeps ongoing records of the reasons why cases are closed. This, however, says little about whether a dispute was resolved. An ongoing satisfaction survey of clients is carried when a case is closed, but this does not cover satisfaction with outcomes. Team performance targets focus on response times rather than outcomes. For many legal cases, the outcome in terms of resolution of the dispute was not yet know, either because the cases, while protracted, were still very recent, or because it would require further (highly sensitive) investigation to track the perpetrator following transfer or eviction.

4.12 Despite these constraints on the quality of data, it is possible to provide detailed analysis of the nature, costs and outcomes of mediation and legal cases. Details of cases are contained in the appendices; these are presented so as not to be identified with any particular mediation service or local authority.

Cost calculations and assumptions: legal cases

4.13 The methods used had to accommodate the lack of routine time recording in most organisations. Furthermore, in relation to legal action, time monitoring could not be carried out specifically for this project because the lifetime of an average ASBO or eviction case well exceeds that of the project. In local authorities, on the whole, legal costs are not recharged for each case, but are contained within a comprehensive charge to the housing service for all legal or all central services. Different organisational arrangements and degrees of specialisation exist among solicitors, from those who undertake only housing investigation work, to those who cover all housing work, to those charged with the whole range of local authority duties. In most areas without investigation teams, due to other commitments, housing officers and solicitors were able to calculate retrospectively the time spent on only one or two cases.

4.14 On the other hand, mediation services were able to carry out time recording for this project because of the generally shorter life of mediation cases. Costs of mediation cases were then calculated from using time sheets and staff hourly rates based on gross salary costs (where volunteers were used, rates were based on equivalent paid mediator salaries).

4.15 In a number of legal cases, costs were estimated by the research team, because the services' own calculations could not be provided within the time frame of the research. The information from specialist teams is more reliable, due to the detailed time recording in place in such teams for the purposes of calculating overtime. Some teams have also already calculated hourly costs in preparation for the service to be contracted out to other social landlords.

4.16 Table 2 shows the relative degrees of quality of the costings data for legal cases from each source. In general, where time recording data was not available, housing and legal services' costs have been estimated based on similar cases from other areas. Costs for police and other services such as social work have not been estimated, except where records were available of the number of callouts or meetings attended in connection with the case. But these services are likely to have been involved in most if not all cases. Therefore, total costs are likely to be underestimates. In each local authority area, slightly different cost assumptions were made in order to arrive as close to actual costs as possible. Details of assumptions are given with each table in the appendices.

Table 2: Quality of cost data in legal cases